Forget Power: A High OBP, Optimized Pirates Lineup Will Have More Offense

BRADENTON, Fla.“We haven’t won the division. So what’s to say it’s not worth taking a look at something else? We’ve got to explore all the different opportunities.”

The Pirates have been no strangers to embracing advanced statistics and unconventional approaches on the field, in order to gain an extra edge on the competition. They were one of the first teams to embrace defensive shifts to a large degree. The same goes for focusing on strong pitch framing catchers, and placing a high value on catcher defense. Some of the approaches haven’t worked out — the original “no-triples” outfield shifts and the approach where pitchers ignored base runners completely in 2012 — but for the most part, the Pirates have been ahead of the trend and have found ways to add a few extra wins to their total.

This year’s focus is a concept that has been around for several years, but still isn’t widely adopted across the game, simply because it’s not how the game was always played. The Pirates are shuffling their lineup, putting their best hitters at the top, and ignoring speed in the leadoff spot. Andrew McCutchen has been batting second for the last week, while John Jaso has been hitting first overall.

As I said, this isn’t a new approach. It’s been around from Tom Tango for some time. You can find the summary of the idea at this article from 2009, but the quick summary is below.

Another way to look at things is to order the batting slots by the leveraged value of the out. In plain English (sort of), we want to know how costly making an out is by each lineup position, based on the base-out situations they most often find themselves in, and then weighted by how often each lineup spot comes to the plate. Here’s how the lineup spots rank in the importance of avoiding outs:

#1, #4, #2, #5, #3, #6, #7, #8, #9

So, you want your best three hitters to hit in the #1, #4, and #2 spots. Distribute them so OBP is higher in the order and SLG is lower. Then place your fourth and fifth best hitters, with the #5 spot usually seeing the better hitter, unless he’s a high-homerun guy. Then place your four remaining hitters in decreasing order of overall hitting ability, with basestealers ahead of singles hitters. Finally, stop talking like the lineup is a make-or-break decision.

The biggest change here is moving McCutchen to the number two spot from the number three spot. The change was made after the Pirates realized last year that McCutchen came up to the plate 158 times (23% of his plate appearances) with no runners and two outs. That was second, only to Paul Goldschmidt. The move goes against the long-held idea that the best player in your lineup hits third.

“The challenge for me is, for 47 years the baddest dude in the game hit third,” Hurdle said. “And that started to flip a few years ago as you look at it. I think it makes some sense. I’ve got to kind of re-arrange my thinking on it, and what’s best for our team. How do we maximize our run production? How do we find a way to put our most consistent, highest opportunity to be the productive lineup? All these different metrics are giving us different run totals.”

The run totals aren’t going to be massive by optimizing your lineup. You might see 30 extra runs over the course of a season with this approach. But that translates to three extra wins. The Pirates are in a situation in the NL Central where every win counts.

“We haven’t won the division,” Hurdle said. “So what’s to say it’s not worth taking a look at something else? We’ve got to explore all the different opportunities.”

If they can do something as simple as batting McCutchen second to get a few extra wins, then it makes all the sense in the world to do it.

“The reality is we’re always going to have to find small ways to get better, whether it’s how we do things or who’s doing them,” Neal Huntington said.

Huntington pointed out that the Pirates needed to get the best players, aka the guys who get on base the most, the most opportunities. He also mentioned that Starling Marte and Gregory Polanco didn’t run much in front of McCutchen last year, not wanting to run into outs. This takes a huge weapon from their game. And that explains why the Pirates have no concerns about the slower, but high OBP hitting John Jaso batting leadoff.

“It’s not a perfect world,” Hurdle said on Jaso’s lack of speed at the top. “Slowing up guys, I don’t know about that. I know sometimes, guys are hesitant to run when you’ve got a guy who can swing the bat in front of them, because they don’t want to make an out. So there’s the other side of it. I think he can go first to third. I don’t think he’s going to clog up the bases at all. Matter of fact, I think he’s a very solid base runner. I think his speed will play out the way it plays out. So that’s not that big of a concern.”

This is an approach that the Pirates couldn’t really take in previous years. Last year they had a more traditional lineup, with Pedro Alvarez being the typical power hitter in the middle of the group. The team shifted their focus to on-base percentage and fewer strikeouts this year, replacing Alvarez with John Jaso. They did something similar the year before, switching from Russell Martin to Francisco Cervelli, and seeing a drop in power for an increase in OBP.

“Cervelli was the first guy for a full year, if you look at his numbers, that we wanted to make sure we just let him simmer,” Hurdle said of the focus for OBP. “Let him keep doing what he was doing. He showed us a year of it. Kang has the potential, healthy upon his return, on-base percentage was an elevated number. We didn’t really have any other guys at the time that have been .340, .350, .360 guys. You look at .310, .320, and those areas.”

The Pirates now have the potential for a great OBP team. Cervelli has been in the .370 range in each of the last three seasons, although the 2013-14 seasons combined for 223 plate appearances. Jaso was .380 or higher in three of the last four years. McCutchen has been the best by far, topping .400 in each of the last four years. And Kang put up a .355 OBP in his first season in the Majors.

As usual, this approach started in the advanced stats department, with Dan Fox, Mike Fitzgerald, Josh Smith, and others crunching the numbers. This was all based on the existing research, although the Pirates finally had a team where this could work.

“You read some of the really intelligent people that are out there and some of the things they’re talking about, then you try to figure out how does it apply to what you’re actually doing,” Huntington said. “The difference between theory and reality sometimes is light years apart, and other times it becomes relatively close. It’s something we’ve actually talked about for years. As we looked at this group and putting this group in a position to find a little bit better offensive production in a perfect world, at least in the theoretical world, this was something where we came together. Now we’re working to see how it plays out in the real world.”

It’s one thing to have a theory and players who fit that theory. It’s another to have a manager willing to go along with the idea, and take such unconventional approaches with each lineup spot. Clint Hurdle has been very open to these ideas in the past, but it wasn’t always that way.

“I paid attention to the numbers for the most part the year I was fired,” Hurdle said. “They were just starting to trickle in, in Colorado. The year I spent with MLB Network in the back room, it was like free play. It was something that I was kind of leery to go back to, but once I got in, I started talking to the guys, I had no where else to go. I had nothing to do. I said, let me check it out.”

Hurdle started digging into the numbers, and it opened his eyes. He recognized that this was the way the game was heading. He still recognized that the numbers side of things didn’t factor the human element, but that the numbers played an important role which was ignored in the past.

“When I got here, it was a whole new gig for me, with Fox and Fitzgerald, and the programs that had already been put in, defensive metrics in our minor league system,” Hurdle said. “It was a buy-in investment. It didn’t happen overnight.”

Breaking Down the Lineup

The Pirates have been very consistent this week batting John Jaso at the top of the lineup. They’ve also had McCutchen as a fixture at number two. You can pretty much mark it down that this will be the order heading into the season. This order makes sense, per The Book. The leadoff hitter should be one of the best hitters on the team, and without home run power. In terms of OBP, Jaso would be second on the team to McCutchen, but McCutchen has more power. So it makes sense to bat Jaso first, with a bigger power threat and a high OBP guy behind him.

The number three spot isn’t as important as was once thought, so we’ll get back to it.

The number four spot is the best hitter on the team with power. The Pirates don’t have a lot of power on their team, and have been batting Starling Marte in this spot this Spring. It makes a lot of sense, as he has power and is fairly good at getting on base.

The next best hitter should go in the number five spot, which is why it makes sense that Francisco Cervelli has been hitting behind Marte a lot.

The Pirates had David Freese in the number three spot on Wednesday, and this makes sense, considering the lowered importance of the position. You want a good hitter, and a guy with some power, but not your best guy. It will be interesting to see how this is handled when Kang returns. I think you could make an argument for Kang hitting cleanup, which raises the question of where you hit Marte (he’s got a lower OBP, but some power, so third makes sense).

The six through nine spots are arranged in decreasing order, with speed up top. The Pirates had Gregory Polanco batting sixth, ahead of Josh Harrison and Jordy Mercer. The latter guys are arranged in the right order by OBP, and are both are weak in terms of power. Putting Polanco ahead of them allows his speed to play. He can move into scoring position without having to worry about taking the bat out of a good hitter’s hands, and by moving into scoring position, he increases his odds of scoring a run with Harrison and Mercer unlikely to do more than a single or a double.

To recap, the lineup that seems most likely:

  1. John Jaso
  2. Andrew McCutchen
  3. David Freese
  4. Starling Marte
  5. Francisco Cervelli
  6. Gregory Polanco
  7. Josh Harrison
  8. Jordy Mercer
  9. Pitcher

Using this year’s ZiPS projections, the Pirates lineup with Freese would amount to 4.617 runs per game with this optimization (I used Gerrit Cole as the pitcher in my scenario, using this lineup tool). When Jung-ho Kang returns, if you do nothing else but swap Freese out for Kang, you get 4.776 runs per game. By comparison, the Pirates averaged 4.302 runs per game in 2015.

So despite the loss of power, this year’s high OBP, highly optimized lineup projects for 51-77 extra runs on the entire season over last year’s totals. Of course, the flaw here is that it won’t be that exact lineup every single night. But it’s easy to see that the lineup this year should lead to more offense than last year’s lineup, and it’s not out of the question to think the difference could be measured in enough runs to equal an improvement of a few wins.

  • I can remember when Brian Downing led off for the Angels.

  • Hard to see. The Pirates had plenty of base runners lasr year and couldn’t drive them in. With less power, what is going to make the difference?

    • “Couldnt drive them in”.

      Not the most accurate narrative. 1st in chances with RISP, 4th in RBIs. So if they did underperform due to more chances, it really wasnt by much.

      They hit for the 7th best average with RISP, and had the best OBP with RISP, but also struck out top 5.

      So it seems like they went for getting better at what they were good at, and cutting the Ks. Pros and cons, but if they lost HRs they gained OBP so you’re somewhat of a wash there.

      If they cut the K rate a few percentage points and Kang isnt a totally different hitter, i dont see any reason they go from a fine team with RISP to poor.

      • There is a perfectly good reason the Pirates could go from good to poor in RISP, and it’s the natural variance in RISP numbers. It has little to do with roster construction and lineup optimization, that fluctuation just happens.

        But going to a high OBP team and cutting the strikeouts should increase the number of RISP situations, which maximizes whatever RISP production the Pirates get. Even if their RISP numbers worsen, the offense will still produce runs if there are a lot of baserunners to drive in. Cutting the Ks also increases the opportunity for hits to happen, since hits can’t happen on strikeouts, and any ball in play has at least some chance.

        I do buy this as an organizational philosophy and not just a cost-cutting measure, though. It’s pretty clear from the way they’ve been drafting that they value hit tool above power. You can also see it in how they’ve handled Bell’s development, trying to foster growth in power, but stopping short of making extreme changes to turn him into a fly ball hitter. I suspect this is because they value his amazing contact and on-base skill, and don’t want to jeopardize it.

        Basically, what it looks like to me, is that the Pirates have identified a thing they can control offensively. They can’t control RISP numbers, they can’t control how many home runs will be hit while people are on base. What they can control, though, is how many guys are good at getting on base, and thus how many baserunners and run-scoring opportunities the team generates throughout the year.

        • Well said.

        • Only thing, has anyone actually shown this lineup to be “high OBP”?

          It sounds great, but is it?

          • It’s better than last year’s, which I guess is the comparison point. Harrison is roughly a wash with Walker in OBP, and Jaso is far superior to Alvarez in that regard. Cervelli may take a step back, but there’s a good deal of OBP upside for both Polanco and Mercer, neither or whom are likely to be worse in that respect than they were last year.

            So overall, compared to last year’s lineup, I think it’s pretty fair to call this one “high OBP.” And given the Pirates were a top-10 OBP team last year, I think it’s also fair to expect to be able to fairly call them a high OBP team this year, especially since only five more points in OBP would have put the Bucs second in that category in 2015. I mean, that’s a lot of events to get to second, but top 5 (only a 2-point gain) is within reach from the definite gains of Jaso, even with a Cervelli step back. But how high do they have to rank to qualify as high OBP? If top 5 is good enough, I’d put at absolute worst even odds on the Pirates being there.

  • This is a good overview of the concepts, though I always wonder how much intent there is this supposed shift from power strikeouts to on-base. Was there a grand strategy or was it simply getting rid of higher priced player whose value doesn’t match his salary in favor a lower cost option whose value is much closer to cost.

    Also, I cannot take the 50-80 run estimate seriously. Using the Zips projections and the listed lineup (Kang’s projection 3rd instead of Freese’s) the Lineup Tool spits out 4.337 runs/game. Using the Zips projections and a lineup similar to what the Pirates used last season.

    Polanco, Harrison, McCutchen, Marte, Kang, Jaso, Cervelli, Mercer, pitcher, you get 4.220 runs/game, about 19 runs over the length of a season. A 19 run change is still pretty massive, that is like replacing McCutchen with a league average hitter for 2 months, most of that change come obviously not from McCutchen’s batting 2nd instead of 3rd but having Jaso and Kang higher in the order.

    Would the 2016 lineup look like this without this optimization push by the Pirates I don’t know, but I think this is a more correct comparison than comparing last season to the upcoming projected season.

  • Haydn Thomas
    March 25, 2016 1:11 am

    Reading the numbers crunchers a while back, they convinced me that a teams record in one run games is pretty much a crapshoot. The same guys that win two thirds one year will just as likely lose two thirds the next. So while I think we will be a better team this year, I worry that the record in one run games is unsustainable. Please tell me I’m wrong!

    • If I remember correctly, bullpen strength is actually a fairly significant factor here, or at least some new evidence seems to support that bullpen strength matters for performance in one-run games.

      • This seems to be the obvious biggest reason for the Pirates beating projections.

        Still though, bullpens aren’t exactly predictable.

        • No, which is a shame, but they have high marginal value for a contending team, which is likely why the Pirates were fine with investing as much in theirs as they did.

          • I credit Huntington for adapting. This obviously wasn’t their intent from the start, but he found a way to grab value when he saw it. Well done.

  • I prefer:

    • It’s funny how diametrically opposed I am to your line-up mate, which is the best reason to assume it is ok 🙂 :)… Just not for me….

      • Could be interesting to post a line-up when Jaso isn’t starting due to a Lefty starter, they still exist…. I’ll take a stab…

        Morse/Rogers (who knows)
        Freese (7/8 i could be swayed for sure)

        • Well posting alone, been there before and it is 9:30pm in Singapore so time to get motivated and go out for the night… Look forward to reading posts about Cutch batting 2nd for the next 7 months as it doesn’t look like that is going to change. But the discourse will certainly change based on the results, of course… best 🙂

          • I have never been to Singapore, what is it like? For purposes of comparison I have been to: Shanghai, Nanjing,
            Shenzen, Sydney, Brisbane, Auckland, London, Edinburgh, Glasgow.

        • Here’s my effort against LHP:


          Same reasoning as above, except against LHP. Also, last year Polanco exhibited a strong platoon split. I want to give him time to grow out of it, not platoon him, to take advantage of his defensive prowess. But placing him at 9 limits his AB and makes him a defacto pseudo leadoff hitter as the lineup turns over, placing “speed at the top” as a tip of the cap to the classic lineup construction.

      • There are several thoughts that go into making this lineup construction:
        1) Put high OBP against RHP at the top of the order ahead of the batters most likely to drive them in: Cutch, Marte.
        2) Place speed (Marte and Polanco) ahead of the batters most likely to benefit from seeing a higher percentage of fastballs (Harrison, Mercer)
        3) Keep Marte and Polanco in positions where they are expected to produce power, to encourage them to adjust their swings to make it happen. “As a man thinks in his heart, so is he…”.

        Polanco in particular needs to abandon his “chop down on the ball” swing, trying to hit the ball to the left side of the infield to get aboard using his speed, ala Matty Alou. My guess is that some well meaning, but ultimately harmful coach taught him this when he was a hopelessly skinny kid who didn’t project for any power. Now he has grown into a beast, and he needs to be swinging like Willie Stargell, not like Matty Alou. Polanco should be sentenced to daily 15 minute sessions of watching Ted Williams swing (with occasional Babe Ruth clips for diversion), observing everything from hand position to swing path, etc.. He should then be required to write the definitive paper on the left handed power swing and be told to go out and execute the same.

  • One other thing I just want to mention. I know it’s spring training, but does hitting a .185 Jaso with a .556 OPS leadoff really make sense when he also has no speed? Oh….and Cutch has 5 homeruns, that’s really going to be optimized hitting 2nd. Okay, sarcasm over. I’m done on this lineup topic, it pisses me off

  • All in favor of the nerd stats, but if Cutch came up with no on and two outs 23% of the time as the #3 hitter…what are the odds he’ll be appreciably better following the #9 and #1?

    Even assuming Jaso is better than his career average and has an OBP of…say .380…that’s 62% of the time he won’t be on base. Since it’s abundantly likely he won’t be on base…and he’s good at it…how likely is it the pitcher’s spot will have made it to first to keep Cutch from being in such a similar statistical position?

    • You’re assuming he’ll have a ton of situations where an inning starts with the number nine hitter.

      Last year Walker and Marte had 68 of these situations each, and they pretty much split the #2 spot. So other instances could have come in other spots for each guy. Polanco was mostly leadoff and number two, and had 83. You could probably see McCutchen’s “no on, two out”(NOTO) totals cut in half from last year.

      • I’m not opposed to the out-of-the-box thinking, but I’m still trying to wrap my brain around how much more effective Cutch will be following Locke/Jaso as opposed to, say, Polanco/Marte.

        It’ll be interesting to see the experiment play out. I would think in the first inning and the latter frames (when a PH is used for #9) there should be an uptick. And I just wonder if this will be enough to counterbalance the weakening with the higher batting placement in innings 2-6.

        I’m not sold one way or the other, on one hand it’ll be great to have Cutch nab another 20ish PAs…conversely, this may be as gimmick-y as batting the pitcher 8th.

        • There is an argument for the pitcher batting eighth so that you stack the lineup and avoid this situation. But the results are minimal, with about 2 extra runs per year. So it’s not a huge thing.

          • A lot of this will just be hard to quantify, though. If Cutch bats second and has a monster season, is it the placement in the order or not dealing with the nagging injuries? If his numbers are the same, would they have been better or worse if he were lower in the order?

            I know stat-heads don’t like to play with the counting numbers, but, I suppose what gives me the most concern is the Angels batted Trout in the 2 spot for much of last season…and his .991 OPS and 41 HRs led to 104 runs and 90 RBI…and he rarely had a pitcher hitting two spots ahead. Cutch, conversely, put up 91 runs and 96 RBI while hitting 18 less HRs and having an OPS > 10% lower.

            • Wow, actually, after really looking at Trout’s totals from last season…

            • The non-Trout Angels also happen to be terrible, so that’s a factor.

              • True…very true…but, by the image below, I’m not seeing a bump in Trout’s numbers when he bats second as opposed to third…and that’s with have an actual batter (though not necessarily a good one) in the 9 spot instead of a pitcher.

                I’m all for the Pirates trying it out and hopefully getting a little edge…I’m just not sold.

                • He had five more at bats in the 2 hole, drove in two fewer and scored two more runs, and his rate stats were worse across the board. Most of that was probably due to the 40-point BABIP discrepancy, which you would expect to be a pretty noisy factor. I’d say his overall run-producing value was better suited for the 2 versus the 3 spot from what I see here. His actual run production was effectively a wash between the two spots *despite* poorer batted ball luck in the 2-hole.

                  • And that brings up two things for me.

                    First, if it is a wash, then I don’t see an advantage or disadvantage to moving him up outside of the fact that it should allow for more PAs. Of course, if you have a choice between 680 or 700 PAs from Mike Trout, sure, on the surface you take the greater number…no brainer. Same for Cutch. But, considering they’ll be batting closer to the bottom of the order, will the increased PAs result in a it net positive difference? In Trout’s case, batting closer to the bottom shouldn’t affect him too negatively as they employ the DH. With Cutch, he’ll be batting closer to the pitcher’s spot…or, a pretty much guaranteed out. Moving your best hitter to the #2 slot sound great, but I go back to the Rickey Henderson quote (I think it was him, anyway…) when asked if he liked being a leadoff hitter he responded: “yeah, but I only do it once a game.”

                    The other quibble I have…and the stat-folks are going to love this…is the ‘because his BABIP was lower, he just had bad luck…”

                    Meh, I don’t know. Could be, but one would assume the ‘luck’ factor is something that would vanish with the growth of the data set. Trout’s 350ish PAs (from both the 2 and 3 slot) isn’t a huge number, but it’s large enough that ‘luck’ should be greatly minimized. Yet, he still had a -40 BABIP in the 2 spot. Maybe it was ‘luck’, but I wouldn’t immediately attribute it to that. My first thought would be: is he being pitched differently and is that restricting his output? I don’t have an answer to that, but, I could envision that being in the 2 slot and batting closer to lesser hitters in the order may mean that he’s coming up in more situations where pitchers can be more selective with their offerings.

                    • It’s not bad luck in the second spot necessarily. It could also be good luck in the third. I mean, .365 is a really high BABIP, even for a player who profiles well for that skill like Trout does.

                      As for whether he was pitched differently, I doubt it. Everyone knows Mike Trout is the guy who’s going to do the most damage. Everyone’s going to try to stay away from him as much as possible as a result, try to pitch him fine. That won’t change no matter where he appears in the lineup.

                      And the pitcher’s spot will not affect McCutchen as much as folks seem to think. Pitchers hit twice to three times a game, some of those times they’ll be sacrificing, others they’ll be the last out. I just don’t see there being a high enough number of events to hurt Cutch. And the situational numbers from last year kind of bear that out. This is why batting the pitcher 8th only has a two-run impact over the course of a full season. The pitcher doesn’t actually hurt the top of the order that much.

                    • I’ll fully concede you may be correct.

                      I’m all in favor of putting out the best lineup/batting order. This is an intriguing idea, but there are so many intangibles at work that whether Cutch’s numbers improve, stay the same, or decrease by being moved up…there are good arguments that can be made that it had everything to do with the switch…or nothing.

                      That’s why I find Trout’s results so interesting. His PAs were basically even between the 2 slots for an entire season…and so were the results. Having two equal data sets from the same player and in the same season, I think, is informative.

                      I also agree, the pitcher’s slot isn’t going to have a great effect later in the game, but it will have a slight one in innings 2-6…that’s why I think it’s a balancing act…yes, the team will get more PAs from Cutch, but will it counterbalance the weaker hitters in front of him? No idea.

                      Overall, I don’t think this is going to make a huge difference, but, if it squeezes out a couple more wins over the course of the season…why not?

                    • I guess what I’m trying to get at with Trout’s numbers is that, when a positive outcome of his at bat occurred, it was more likely to produce a run in the 2 spot versus the 3. This could be coincidence, but it could also be a function of lineup optimization.

                      Consider also: Trout walked *more often* and also struck out *more often* in the 3 spot. The differences are too small to draw a real conclusion, but it seems to suggest he wasn’t simply being pitched around more often in the 2, since both higher walk and higher strikeout rates do imply pitchers staying more on the fringes of the zone. He also got hit by more pitches in the 3 spot, for whatever that’s worth.

                    • I get what you’re saying…at least I think I do…

                      It seems you’re making the analytics argument that: “If his BABIP was equal among his two hitting slots, his #2 stats would’ve been better than #3.” I don’t disagree that that’s what seems to be the case.

                      However, I just don’t share the perspective. I’m not viewing it as what would’ve been if the numbers were normalized, but what they actually were since the opportunities were equal…and, aside for insignificant variances, there was no difference.

                      I’m not sure what the Angels plan is for Trout this season. If they’re going to continue to split his PAs between the 2/3 slots, I’ll definitely be keeping an eye on it to see what results. It’d be interesting to see what the data looks like if he does this for another season or two. If anything, having one player fill both holes takes away the side arguments of: “Well, he was healthier this year,” or “if he did that well, what we have done if he stayed put in batting order,” etc.

                    • I don’t think anyone has ever claimed that lineup optimization is even likely to produce a couple wins. Not even close.

                      A couple wins, just by moving a couple guys around in the batting order, is ridiculously huge. Think about it.

                      This is why those “lineup optimization” estimates in Tim’s article are ridiculous.

                    • I sort of choked on the +77 runs. Not going to lie.

                      But if it can snag 5-10, that could be the difference in a couple of games.

                      But, even if the number of runs was something like 20…meh, they wouldn’t all come when you needed them. An extra run when you’re up or down 7-2 really doesn’t matter.

                      But it still falls back to the point, if the difference is so slight…it’s almost noise…and how do you isolate its effect in a dynamic system?

                    • Couldn’t agree more.

  • In the 1st inning Cutch will need Jaso to get on base. Let’s say Jaso makes an out and the Bucs get one base runner in each of the 1st 2nd innings. Now Cutch has the pitcher and Jaso ahead of him. It seems they’re just as likely as not to be having him still come up with 2 outs and nobody on. I think you need to have one of your better hitters batting 8th to try to avoid having the pitcher leading off an inning. This might also have more opportunities for the pitchers bunt the runner to set up for Jaso and Cutch. Wait, I forgot that Pirates pitchers can’t bunt.

  • I have not read The Book. And I support a data-driven decision on the lineup. However, I am having a hard time buying that the #3 spot is less important than both #2 and #4.

    I sorta think that if you put your best hitter #2, then collect a bunch of seasons of data, suddenly you’ll find that the #3 spot is now more important than the #4 spot. Because if the analysis is based on what has happened in the past, then this analysis is probably telling us that teams haven’t been putting enough good hitters in the #2 spot, and probably not that the #3 spot is this weird vortex at the top of the order that doesn’t matter as much as the spots around it. If that’s true, then why isn’t the #7 spot more important than the #6 spot?

    What I would love to know is what happens if you put all the players into a simulator and try out the 120 (that is , 5! ) different combinations for the top 5 hitters to see which gives you the most runs on average.

  • The part that keeps really aggrevating me is this point that keeps getting glossed over WRONGLY- “The number four spot is the best hitter on the team with power.” Everyone in the whole world knows this is Cutch. So why is he not hitting 4th? Until someone gives me a real answer that makes sense, i’m going to continue to be aggregaved with this whole approach, it isn’t optimizing shit if you don’t put him where he needs to be. The whole lineup makes more sense with Cutch at 4. Jaso, Cervelli, Marte, Cutch “by optimization rules” makes a TON more sense. Then you go Kang 5, Harrison 6, Polanco 7 (or swith these two) and Mercer 8. It’s a MUCH MUCH MUCH better lineup against righties at least

    • If you bat him fourth you get ~40 less plate appearances in the year. That is one less HR and 6-8 less RBIs…

      • Bruce- did you read about the theory of lineup optimization? Whether you subscribe to “normal” baseball lineup construction or “optimized” NEITHER suggest that his best place in the lineup is 2nd, that is my only point. I’d rather have Cutch knocking in people, than hoping other people knock him in. The power hitter whom hits for average- hits 4th even in lineup optimization, who cares he gets less at bats.

      • No….it’s not, because in those 40 less plate appearances he will still bat with the same relative level of people on base due to the spot in the lineup. There is a reason why #2 hitters don’t get as many RBI’s as #4 hitters. That’s why.

      • Cutch will basically be hitting with Mercer, the Pitcher, and Jaso (when a righty is on the mound) in front of him all game after the first inning, and only Jaso in front of him in the 1st inning. Do you really think that’s going to equal 6 more RBI’s rather than hitting after Jaso, Cervelli, and Polanco/Harrison/Marte just because he has 40 more plate appearances? If you think that, you haven’t been watching much baseball over all these years that you’ve been complaining about us never winning anything.

    • I think it’s because of the extra plate appearances.

      Also, if you put the lineup and ZiPS projection into the optimizer tool I linked to, every top lineup has Jaso first and Cutch second.

      • That’s actually pretty interesting, I appreciate that. I don’t expect anyone to know this, but how do we know the optimizer tool is making accurate assumptions? I guess that Cutches additional value add at #2 is more over the next best option at that spot vs. Cutches value at #4 over the next best option at that spot. SO….even though Cutch fits better at #4, he optimizes our lineup better at #2 because our personnel isn’t strong enough for another player to maximize the #2 and maxmize the #4 spots properly.

        • You enter OBP and SLG for each player, and it calculates out the best lineups.

        • I tried this tool out and put in Zips projections for players in 2016, and basically none of the lineup suggestions had Cutch 4th (looks like the tool lists the top 30 lineup possibilities based on the Runs per game they would be projected to produce). Most have him batting second with a large portion also batting him leadoff.

          Starling and Josh are the two guys neck and neck for cleanup on these lists, basically Marte edging Josh out in the 4th spot and Josh sliding into 5th.

          The other interesting takeaway from this tool is it bats the pitcher 8th in all of these lineups, with Cervelli batting 9th to basically be another option to get on base in front of Cutch in the second spot.

          *EDIT* That was with Freese, when you add Kang back into the mix it actually moves Josh to bat 3rd and moves Kang to 5th

          • The thing I wonder is this: if you don’t follow the optimized lineup and try to pick and choose whom you want to put in what spots, doesn’t that take away from the value add? Meaning, is Cutch still a fit for #2 if the pitcher still hits 9th for example.

            • Yeah its a very good point for sure, and to me I actually always thought that the lineups where the pitcher hit 8th were ones optimizing for the #2 guy (like Joe Madden lineups)

    • If you focus on “best hitter with power,” you gloss over “most well-rounded hitter,” the designation for the second spot. So yeah, Cutch is the best hitter with power, but he’s also the most well-rounded hitter. Over the last four seasons, McCutchen has the 3rd best OBP and the 30th best ISO in baseball. Yes, he’s an elite power hitter, but the skill which really stands out is his absurd on-base ability. In terms of maximizing Cutch’s RBIs, sure, batting him 4th would make sense, but maximizing the team’s run scoring potential? That’s a different matter.

      There are a number of capable hitters in the Bucs’ lineup, but Jaso is the only other one who really appraoches McCutchen in on-base skill. Batting McCutchen second won’t eliminate him as an RBI guy, but it will definitely maximize him as a run scorer. And when it comes down to it, both of those matter in terms of the number of runs the Pirates score.

      But there’s also the matter of literally no one else in the lineup profiling well to slot in at #2. Marte doesn’t walk enough, Jaso and Cervelli don’t have enough power, Kang’s on-base skill falls just shy, and Harrison, Freese, and Mercer simply aren’t good enough hitters. But Marte and Kang each profile just fine for the best [remaining] hitter with power role for #4.

      This is the balancing act, though.

      • dark- that is fair- But here’s the biggest thing: #4 lineup spot is listed by lineup optimization of being at least slightly more important than #2, and Cutch is the living breathing epitome of “best hitter with power”. If the #4 slot is even a tiny bit more important, and Cutch fits that profile to a tee, why would you use the A “well rounded hitter” idea to put him at #2? To me, “best hitter with power > most well-rounded hitter” I guess maybe there is some perception and lack of clarity in those statments that they themselves should alter to get rid of the merkiness of their system

  • Where do you think Josh Bell would fit into the lineup when he comes up? Would he replace Jaso as the lead off hitter due to his good OBP and the fact that his power is still developing?

    • 10th. Just kidding. He’s a rookie, he isn’t going to come up and hit .360 OBP. He’ll hit somewhere he can see good pitches, but don’t expect that to be in 2016

    • domdidominic
      March 25, 2016 7:09 am

      Will be interesting to see where they bat him in Indy. He should be the best hitting on the club, so #2. If he is preforming well, he will be getting those Jaso AB’s (when Jaso is not in the OF), so does Indy bat him #1?

  • I’m all for the new analytics but sometimes it forgets to factor in the human adjustments that can be made by the other team. If our 5th best hitter is batting behind cutch pitchers aren’t going to be exactly pounding the zone giving him pitches to hit. I know he’s got good discipline but sooner or later he’s going to start expanding the zone. I still think cutch second but he needs some combination of marte/kang/cervelli behind him in the 3 & 4 spot

    • Why is he going to suddenly expand the zone? I get the assumption that he sees less hittable pitches (though at this point, that seems odd since anyone with a brain already should be doing what they can to not give him anything) but i dont see why he starts expanding.

      He’s posting really good OBP numbers and walks at an above average rate. If a team really wants to have the first two men of the game get on, im all for it. Freese isnt going to tank the inning all by himself so often that its cringe worthy, and he’s there for 2-4 weeks before we re visit who is where.

      • I do think it makes sense to have a high-average guy behind Cutch in this scenario, though. In all the non-first inning turns of the lineup, I could imagine plenty of situations where Cutch gets up with two outs and a guy on base, and sees nothing to hit.

        Absolutely agree that nobody is “protecting” one of the top three hitters in the league, as in making pitchers through Cutch fastballs in the zone, but a guy who could put up a .280+ average can drive runs in without even needing power in this situation.

        • I think the plan is simply Marte there when everything gets healthy. Marte isnt a big OBP guy, but has become good enough with his OBP his fits at the 3 in that scenario fine. He’ll drive in runs, have pop, and not be .320 OBP like.

          Kang just needs to be healthy. 3-4 weeks of Freese at the 3 is good enough for me, mostly because i value Polanco’s speed enough that i want him lower and able to simply not worry about who is hitting if he’s on 1B.

          • I’m still holding out for a mid-20s power surge from Marte and love him in that 4-hole; even trading 5% of all those ground balls for flies turns him into a monster like we saw in the second half of 2014.

            But pragmatically speaking, I do think you have the right idea.

        • The simulations suggest high power is more valuable than high average in the 3-spot, whereas the 5 demands a better on-base guy. It may be the upside of extra base hits with two high-OBP guys in front of you outweighs the downside of fewer hits.

          Even so, I think I’d rather see someone other than Freese there. I just can’t figure out who. Polanco would be great if he came into his own, obviously, but that’s not a sure thing. Harrison and Cervelli don’t really have power. Mercer’s the worst hitter among the regulars. Freese might be the best option there despite his flaws.

          When Kang gets back, though, he or Marte make a ton of sense at 3. I’d lean toward Marte, since Kang seems to be the better hitter to slot in at 4. And if Polanco does come into his own this year and becomes a really good all-around hitter, and especially if his OBP skill develops, maybe he earns the leadoff spot and pushes Jaso to 5th, or Polanco takes over at 5th. Then you can have Harrison and Cervelli follow him.

          Basically, a healthy Kang and a step forward by Polanco will make this lineup a ton of fun to watch, and possibly one of baseball’s best.

          • “The simulations suggest high power is more valuable than high average in the 3-spot…”

            Think about this, though. How much sense does this really make in terms of the Pirates lineup? If the goal is to come up with a hit after Cutch gets on, do you really need anything more than a double to score him? Maybe we’re arguing semantics, but I don’t see doubles as indicative of “high power”.

            • I actually was including doubles. In fact, I was thinking specifically of doubles when I talked about the upside of a power guy being there.

              Until Kang gets back, though, yeah there’s some challenge in getting a power guy in that spot without sacrificing a little bit in terms of the quality of the overall hitter. But that’s also largely because until Kang gets back, there are too few really good hitters to use there.

              • Gotcha.

                Harrison actually led the Pirates in 2B/PA over each of the past two years, fwiw. It’s just that doubles correlate poorly with ISO, which typically is used to denote power.

    • Yes but your high OBP guy is leading off. The intent is if he is on base then Cutch will see more strikes to avoid adding a base runner via a walk. I agree with having Marte 3 and Kang 4th

      • If you have Cutch 2nd, you can either have Marte 3th or 5th and it’s a better fit than 4th. I personally like 5th better because there is noone which has power behind him which would slow down his baserunning. If Marte can get 40+ bags this year, we might actually be able to make this work.

    • I expect Cutch will have to take his walks early on because of Freese hitting behind him. But then Cutch has speed to make them pay by stealing a base and causing Freese to see more fastballs, too.

      • Over/under on the # of double plays Freese hits into batting 3rd behind Jaso and Cutch for the first month

        • Scott Kliesen
          March 25, 2016 9:52 am

          I’ll take the over. He’s certainly going to have his chances to hit into them.

          • Chances will surely drive up his base number, but i dont think he’s particularly prone to hitting into a massive amount of them.

            Okay GB/FB rate, and his ISO is such that he should have enough power to avoid weak contact to a staggeringly high degree. Hits it on the ground, but not super pull heavy and able to spray good enough. If he sit mid to low 30s in his pull ratio, he’ll hit it up the middle and oppo enough to likely not be super easy to shift with.

    • I agree and have stated I absolutely expect us to lead the NL in players left on base this year. If anyone will even give me 5-1 odds on it, I’ll put up $100 right now.

    • Before he had 23% of plate appearances where there was no reason to pitch to him.

      Now? He’ll start 36% of games with a runner on base and no outs with Jaso hitting in front of him. Do you want to pitch around him and start one game per series with two on and no outs, all because the number three hitter is maybe the 5th best on the team?

      And then think about all of the other situations where pitching around him makes no sense because it puts extra runners on with less than two outs.

  • Interesting article for sure.

  • Worst case scenario — Jaso walks, Cutch walks Freese grounds into a 2bl play, Marte, feeling the pressure of hitting 4th strikes out chasing.
    0 Runs
    Best case scenario — Jaso walks, Cutch 2bls him home, Freese grounds out, Marte singles Cutch home then steals 2nd, Cervelli singles him home, J HA or Mercer single & J HA or Mercer & the pitcher make outs. 3 Runs & the lineUp is turned over.

  • The only problem I have with this is David Freese batting third. From watching baseball my whole life it seems like hitting comes in bursts and has a lot to do with momentum(are there any stats to back this up). And I think you’d want one of the better hitters to be in the middle of where one of those hot streaks is likely to occur.

    • Freese has no business hitting 3rd, you are correct. Hopefully right now he’s hitting 3rd to get in the swing of things. Harrison would probably be the best #3 hitter of whom is available given their…….&%*#* approach to #1 and #2

  • There’s a typo in the first paragraph where you said the pitchers ignored the baserunners in 2012…I think you meant to say all of the years.

    • They ignored them completely in 2012.

      Last year, for example, they ignored them in certain counts. If it was two outs, they let them run and didn’t bother, instead focusing on the runner at the plate. It’s an evolving strategy. In 2012, there was no strategy. It was just ignoring them completely in all counts.

      • its a horrible strategy. As i’ve said before, if you can’t throw a good pitch because you are mildly trying to hold a runner on, you shouldn’t be in the major leagues as a pitcher. I will not accept any excuses for neglecting the basic fundamentals of baseball

        • Every single pitcher throws better when they’re not rushing their delivery to the plate. That is a basic fundamental of baseball. And this approach makes sense because of that.

          • No offense Tim, but I didn’t say anything about rushing their delivery to home plate. Every professional pitcher should be able to effectively hold runners on without having to rush their delivery to the plate. You and I both know there are other options to help against the running game. There is a huge mid point between: asking a pitcher to speed up his delivery to aid in defending the running game; and completely ignoring it to the point we aren’t throwing over, aren’t pitching out, aren’t stepping off. That was my point.

            • There are hold times. There are pickoff throws. But if that guy is running, and you’re slower to the plate, you’re not catching him. A huge tactic for every pitcher is speeding up the delivery to the plate. The idea that pitchers can get runners without this method completely is false.

              • Tim, seriously… have been watching baseball your whole life just like me, what are you talking about???? Lets level set. Yes, a faster delivery to the plate does help, every pitcher is faster out of the stretch than the wind up, so that’s a given by itself. Assuming the base runner is going, the following are the main factors in a stolen base:

                1. Lead distance from first base
                2. Jump from first base
                3. Time to get to full speed for the runner (and the speed of the runner in general)
                4. Time to home for pitcher
                5. release time by catcher
                6. accuracy and strength of the throw from catcher.

                You are basically telling me here that if #4 isn’t fast enough- that all other factors are moot. That overgeneralization is ridiculous!

                Lets look back on those situations where we gave up those 2 out stolen bases about 100 times last yea (intentional made up stat). would you like to hazard a guess on how many total throws over we made to first base or how many times we stepped off previous to those stolen bases? I would venture that your original statement was accurate and that we purposely ignored them and weren’t doing much of anything at all to try to stop them from stealing. I would bet everything I own that on those “2 out steals” where we basically didn’t even make a throw- that we stepped off and threw to first base SIGNFICIANTLY LESS than average versus how we play rest of the game.

                That single assertion means it’s not just a matter of speeding up time to the plate, but also Hurdle not believing his professional pitchers can make a good pitch while holding on a runner, and that is quite frankly a horrible strategy.

                If i’m saying anything here that you can prove is wrong or off base…. I welcome it, but if you are just going to argue for the sake of arguing and not engaging in a debate with merit simplybecause you don’t like to admit you could possibly have said something that wasn’t all that well thought out- just keep it to yourself.

                • Where do you think the phrase “steal off the pitcher” comes from?

                  Teams plan steals knowing the runner’s typical overall speed (1-3) and knowing the catcher’s pop times and accuracy (5-6). Those are treated as constants. Then, if they have a pitcher who is slower to the plate (#4, the big variable), and the pitcher + catcher times are greater than the runner, they’ll steal.

                  This is why you see a catcher give up plenty of stolen bases with one pitcher on the mound, and no stolen bases when a pitcher who is faster to the plate gets on the mound.

                  So if a pitcher can slide step and speed his way to the plate, he can reduce the time on his end, making it more likely to catch the runner.

                  • Incidentally, this is why so few people steal on Kershaw, no matter who’s catching him that day. He’s exceptional both at keeping runners from every reaching base and at keeping them from stealing. His catchers are fine, but Kershaw is both quick to the plate and deceptive to the runner, and that makes him really tough to run on.

                    Burnett and Morton were molasses to the plate. Stewart and Cervelli never stood a chance.

                • More drivel, and in this comment, a LOT of it.

                  • Leo- if you want to keep calling comments I spend 20 minutes trying to spell out in detail “drivel” I’m going to lose my patience with you. If you don’t like something I say, then put it in quotes and debate it. Sportsmanship can exist on forums as well as on the baseball field.

      • The catchers then couldn’t throw anyone out anyway.

  • How many times did the #2 hitter come up with two outs and no one on base? I bought into the analysis, but I still think that is something that should be taken into account: you are looking for the difference between one way or the other, so you need to know the other.

    • Polanco had 83 total batting first and second. Marte and Walker each had 68, and they were most often in the 2 spot.

      • I’m more interested in how many times the #4 hitter hit with no runners on and 2 outs because that is the spot which is more important than #2 and fits the “best hitter with power” which is absolutely cutch, not Marte or Kang

        • I don’t think you can boil it down to one number. No one on and 2 outs is just one situation, and it only comes up about 14% of the time. Other higher leverage situations may come up more often that far outweigh the low leverage 0-on, 2-out situation.

          • Arik- that was the point Hurdle keeps making that they wanted to avoid which led to them hitting Cutch 2nd. It’s an incomplete and stupid reason to move him from 3rd to 2nd since you can blame a lot of that on the people hitting in front of him, not his place in the batting order. Again, if you want him to hit with more people on base, in more “valuable” situations, then you drop him to 4th, where he is guaranteed in his first at bat to NEVER hit with bases empty and 2 outs.

            • Not if someone scores.

              • Good math Daniel- You are right, i didn’t think of the ability for a player to score via homerun.

            • Wait, if you’re gonna go around calling other ideas incomplete and stupid ya better come up with a better reason than “because that’s how it’s always been” for having him at #4 over #2.

              • I already have: I’m using the guidelines posted above which clearly state #4 is more important than #2 and the ideas of lineup optimization that I read that state “your best hitter whom has power shoul dbe in the #4 slot) These aren’t my ideas……and I don’t even agree with them, but that’s it. Don’t call me out on this NMR, I already made this clear in numerous posts this week.

  • 51-77 additional runs over last year would place this lineup as the second or third best offensive club in the entire game. Assuming we all agree that doesn’t come close to passing the smell test, you have to wonder just how far off those “projections” are from reality.

    Not sure what’s going on with the Cervelli-Martin comp, but catcher OBP *dropped* from ’14-15′. Pretty damn hard to top Martin’s >.400 OBP, and Cervelli sure didn’t.

    The issue with this “high OBP” lineup is that it’s completely BABIP-dependent. Only two starters project to walk at an above-average rate. Obviously a ton of risk in that sort of profile.

    This club won’t be bad offensively, but let’s pump the breaks a second.

    • Jaso’s an interesting guy though, in that his OBP has been able to survive (in terms of being at least solid) some year dips in BABIP. The few years he BABIP saw drops to .300 he was either able to keep a decent OBP or really good.

      I think a guy like Jaso, even with BABIP being variant, is a good bet to stick at a high OBP and provide a good option for this system. I think Cervelli is certainly more of a risk, but he’s also hitting lower in the order and you can shift some things around if his BABIP tanks his OBP totally.

      Even with worries of BABIP, this is easily a more “high upside” strategy than assuming Harrison or Polanco fits well at the 1.

      • Bingo!

        And why has Jaso managed to maintain good OBP’s despite BABIP fluctuation? Because the dude would rather take a walk than jump on a 2-0 meatball. Walks, walks, walks.

        Also agree that the top of this lineup looks better than previous years.

        • Without a doubt.

          Id be interested to see if they are having Cervelli take that mindset to a new level as well. 2-0 count and you are hunting that walk with Polanco set up behind you thinking about driving in runs. Bit of risk there that assumes Polanco takes a step forward, but thats a key to the season regardless.

          • Polanco has no business batting 5th in any order at this stage in his career, it’s a joke

            • The entire discussion assumes you are thinking in terms of this method of lineup optimization, and i feel like you dont buy into this. Which is fine, but tough to discuss what im getting at if you dont buy into this idea at its base.

              • I don’t buy into it but I understand the concept and agree it can have merit if you have the proper personnel and the discipline to live by it. But damnit, if you are going to do it, do it right. They are doing it half- assed, which is why I would prefer they stick to the “book” if you are going to optimize, then do it correctly. I don’t understand why anyone would expect a partial optimization to even give you partial positive results- there’s no data on that.

                • I dont think you do fully get it if you think we dont have the right personnel for it. They’ve got the high OBP options for the 1 and 5 spots, and with a healthy Kang they have a 3-4 kick that allows Cutch to slide to the 2 without fear of total implosion behind him.

                  They arent partially doing anything. They appear set to throw the high OBP guys at 1,2 and 5. Thats the goal. They throw the next best hitter at 4, and ascending from there. All thats missing is a healthy Kang, allowing Marte to slide to 3 and Kang to 4.

                  Thats what you want. Good to great OBP options at 1-2-4-5, Marte with enough power to allow good enough OBP at 3, and then the obvious 6-7-8. But right now, as is, they have solid options for 1-2-5 and 4.

                  • If we both know that the best place for Cutch is #4, “the best hitter whom also has power”, the best #2 hitter would be your best non-power hitter as per the guidelines on optimization that I read multiple places in the last week. If we are using Cutch to optimize our lineup in a spot which isn’t the best for him based on true lineup optimization( #2 vs. #4) based on his skill set, it means we don’t have the proper #2 hitter on our roster to fully implement the strategy correctly. I 100% understand it.

                    • I may be wrong, but the way I understood it was that your best OBP guy goes in the #2 slot, regardless of power or anything else. That puts Cutch second,
                      I think if Kang returns to anything close to what he was last season, then this makes much more sense, as he can slide into the #4 slot. He is the only person on the roster that can actually claim that he has power.

                    • That isn’t what i read, sorry.

            • One thing id say purely on the notion that Polanco cant hit 5th is that its a double edge sword.

              If you throw him 7th because his base stats (ignoring things like hit velocity) arent good enough, it gets harder for him to see quality pitches when Harrison or Mercer are behind him. No one throws a fastball to Polanco if they can get Mercer next.

              So his OBP isnt good enough to hit early in the order, but burying him does him no favors in seeing hittable stuff. 6th makes sense, both in the article’s proposed system and traditional logic. But so would 5th in you have any faith in his exit velocity showing that he is crushing the ball as often as any player, and is about to take a step forward in terms of consistency.

    • Injuries and rest days (especially the frequency with which the Pirates use them, and the Jaso platoon factor) are likely to dip that number closer to about the 30 the model touts, which puts them in the vicinity of the Royals’ offensive output from last year, which, stylistically at least, is probably the best comp to the Pirates lineup. They’d only slide up a couple spots (the Pirates having been ranked 11th last season) to 8th or 9th, all else equal.

      So for me it still passes the smell test. I’ve suspected this lineup would be better than last year’s at producing runs since we replaced Alvarez with Jaso, and the lineup optimization makes me more sure of it. This result is consistent with that.

      • I mean, you’re certainly free to that opinion, but to me *magnitude* of a projection sure goes a long way to this discussion.

        I’d have no problem if the theory posited that this Pirate club is likely to score more runs than in 2015. But that’s not all the article claimed.

        • It is.

          • Oh please.

            Tim, it is literally in your own article. You wouldn’t have included it if you didn’t believe it supported a claim you were trying to make.

            • To be clear: do you think I’m saying they’ll be 51-77 runs better? Because I thought I made it clear that this was only if the lineup was in every game, which it won’t be.

              All I concluded with was that they’ll be better than 2015, which is something you agreed with.

              • I think they’d be better than 2015 no matter what they did with the lineup, even before they got Freese, but especially now if they start Kang at SS. Without Mercer, the other 8 each project (Steamer) to a wRC+ > 100. And I think it’s light on Kang at 108.

                The only obvious regression candidate is Cervelli, everyone else should give the same or better production than last year.

          • Also, please don’t mistake this one point for thinking the overall article was poor. It’s very good.

            I just don’t believe it supports a claim of the magnitude of improvement you have repeatedly written you believe will happen.

      • Remember that each lineup the pirates trot out can be optimized as well. So while they will rotate players, you aren’t talking about completely overhauling the lineup every Sunday meaning I don’t think it is fair to take Tim’s range and go all the way down to 30.

        • Sure you can “optimize” any lineup, but you most certainly cannot pull OBP’s like Jaso’s and Cutch’s out of a hat, which is the entire point of this optimization.

          • The problem here is that there is no actual possible way of proving whether a lineup change causes more runs or that the players had a better year than 2015- at least i don’t think there is…?!?!? – well I guess we can look at our Team OPS vs. our runs scored as a ratio and if our ratio because “tighter” this year then it worked? Am I thinking about this correctly?

            • I think you’re correct to be skeptical of the above article, at least the part about magnitude of additional runs produced by optimization. The same saber minds that talk about lineup optimization also say that lineup construction has a marginal effect all together. 50-70 additional runs represent a *huge* amount of untapped value. If that was even in the realm of possibility, *every* team would be doing it.

              I think this matters because of course the magnitude of any benefit or cost is going to determine whether or not one thinks they’re good ideas.

              I don’t think there’s any way to prove definitely how much a change like this will work. Simply too many variables to isolate in order to get a strong signal one way or another. If you’re asking me, I’d look at it by situation. How many times are you getting your best hitters up with guys on base. What happens from there really doesn’t matter in terms of optimization, if you ask me. It’s up to the hitters themselves to produce at that point.

    • I was looking at Martin’s career. He had a .400 OBP in 2014, but .327 and .329 in the surrounding years. His last five years have been .324, .311, .327, .402, .329. It didn’t seem right to call him a high-OBP guy based on that year.

      As for walks, Jaso has a 12.6% career rate. Cervelli is 8.5%. McCutchen is 12.1%. League average last year was 7.7%. It’s hardly BABIP dependent with these guys.

      • Jaso and Cutch are the two above average guys I mentioned.

        Cervelli, with only an average walk rate, *absolutely* depends on BABIP to remain a high OBP player. When that >.350 BABIP eventually regresses, and it’s only a matter of when, he’ll have to significantly increase his contact rate to come close to maintaining an OBP over .350. Just comes down to math at that point.

        The rest of the lineup projects to walk even less than Cervelli without striking out appreciably less (Harrison, Polanco), if at all (Kang, Marte), meaning they’ll absolutely have to run high BABIP’s to be much above average.

        • Cervelli is above average. He was 9% last year and 8.5% in his career. We’ve had the discussion many times about the BABIP. You’re ignoring the reasoning behind the high numbers.

          • No, you’re ignoring the *cause* of his high BABIP, and why that’s simply not sustainable.

            I’m not making this up out of nowhere.

            “Francisco Cervelli was another significant offensive overperformer in 2015. Sure, he lacks glaring weaknesses, which in and of itself is a positive, but his skill set is incongruous with a 113 Contact Score. The low pop up rate would appear to be real, and might be his single largest offensive positive, but the extremely low grounder authority is a bit of a red flag, especially considering his high grounder rate. I’m not expecting any 100+ OPS+ seasons from Cervelli in the near term, either.”


            Cervelli simply does not have the speed nor batted ball profile to support what we know causes sustainable extreme BABIP’s, a la Starling Marte.

            • We’ve got a stats vs scouting debate here.

              The Pirates have taken the same approach with both catchers. Use the opposite field. Focus on line drives and ground balls, rather than trying for homers. Aim for singles and doubles. Shorten up your swing and treat everything like it’s a two strike count.

              Stewart came here and his BABIP immediately shot up. Cervelli came here and maintained a high BABIP over a full season. Both had the same approach.

              I’ve talked with both of them about the changes they made immediately after arriving here. I buy it, based on the results, and because it makes sense. You don’t. We’ve had this discussion many times, and we’ll have to agree to disagree.

              • If you can prove you’ve found a sustainable way to maintain extreme BABIP’s with substandard contact and speed profile then you’re missing out on making a boat load of money somewhere.

    • I think we have a deceptively strong offense and will be better than last year when we finished 11th in MLB in runs scored (which is more impressive than it might seem when you consider the only NL teams above us were COL and ARI with their high run scoring environments and WSH, which only scored 6 more runs than we did).

      However, the 51-77 clearly overstates the improvement because we can’t put the optimal lineup out there every game. There are of course games when our prototypical #2 hitter won’t be available, for example.

  • Exciting to see any team buy into this idea more fully, if only from a practice vs theory test case standpoint. Be really interesting to see how the lineup meshes once Kang returns and hopefully slides into the the 4th spot (assuming they are really buying in and feel Kang is better suited than Marte for the more crucial spot).

    • Yes these are exciting times. Need to keep finding an “edge” every year to compensate for MLB economics.

      I am so down with this management team.

    • I would be shocked to see Kang come back strong. May take a whole year to fully recover from That injury.

      • I think its premature to get that reaching about the situation. While certainly not ruling that out, i dont think its really worthwhile to assume that at this point.

        I think its unwise to assume he’ll regain form the moment he plays, and unwise to assume it’ll take that many months for him to be a solid hitter again.

    • Cutch, Marte, Polanco, Kang, and Jay Hay are a pretty good 1-5. Only problem 8 hitters have to see if these young guys like Hanson, Bell and who ever the 4th OF is can handle major league pitching.

      • So long as you totally ignore J Hay’s OBP and walk rate, then yeah he’s great at the top of the order.

        Otherwise he’s a 7th hitter all day. And a good one for that slot.

  • hate freese batting third, he is a 7th place hitter. move polanco or cervelli to bat third vs right handed starters, the other will bat 5th.

  • I wonder how many more runs our rotation projects to give up per game.

    • Think last year it was .6 or .7 per game, probably will be lower w out Locke and Morton.