Exactly four years ago yesterday, Jeff Sackmann wrote an article in the Hardball Times taking a look at what should be expected from each rotation spot. I had read Sackmann’s article before, but stumbled on to it again last week when I was looking for information on the actual results for #1-5 starters. I was looking for that information, because like Sackmann at the time, I felt that people tend to overestimate how good a pitcher should be.
Two weeks ago I gave a preview of the Pirates Prospects 2011 Prospect Guide, putting up the scouting report of Rudy Owens. In the report I mentioned that Owens has the upside of a #2-3 starter, which drew some questions. Some of those questions surrounded my credibility making such a statement. All I can say to that is that I’ve seen Owens over half a dozen times in the last two years, I’ve talked to scouts, coaches, and players about him (both in the Pirates’ organization and outside of the organization), and I’ve even talked to Owens himself about what he’s been working on, some of which has made it on to the site, and some of which has not.
But this article isn’t about my evaluation skills. It also isn’t about whether Owens has #2-3 upside (key word being upside, in that this is his max potential), mostly because I don’t think that’s the issue here. When I say Owens could be a #2-3 starter, I get the feeling that the people saying he’s a #4-5 starter have an inflated view of what a #4-5 starter actually is.
Jeff Sackmann studied the 2006 season to find the results for each spot in the rotation, using the following method:
For the purposes of this article, it’s necessary to define exactly what a #1 starter (or #2, or #3) is. To keep things as simple as possible, I used ERA as a measure of pitching ability. I also figured that each rotation spot accounts for 32 starts. On many teams, the #1 guy isn’t the same for the whole season. For example, let’s look at the 2006 Twins. Here are all of the pitchers who made more than one start for Minnesota last year:Starter GS ERA Liriano 16 2.16 Santana 34 2.77 Bonser 18 4.22 Radke 28 4.32 Garza 9 5.76 Silva 31 5.94 Baker 16 6.37 Lohse 8 7.07
By ERA, Francisco Liriano was the best of these guys, but he only made 16 starts. So, he made half of the “#1 starter” starts. Since Johan Santana is next in line, I assigned 16 of his starts to round out a composite #1 starter. Thus, the Twins #1 starter was half Santana, half Liriano. Santana’s remaining 18 starts were assigned to the composite #2 starter.
Intuitively speaking, that distribution is a reflection of the fact that, while Liriano was in the rotation, Santana was #2. When Liriano was in the bullpen or on the disabled list, Santana was #1.
Sackmann discovered that the rotation spots broke down as follows:
Lg #1 #2 #3 #4 #5 MLB 3.60 4.14 4.58 5.10 6.24 AL 3.70 4.24 4.58 5.09 6.22 NL 3.51 4.04 4.57 5.11 6.26
Yesterday I had a poll, asking for your expectations for each rotation spot. The results were much higher than Sackmann’s research of the 2006 season:
#1 - 36.15% of voters picked a 3.10 ERA as the expectations for a #1 starter. 67.64% had a #1 starter as either a 2.84 and 3.10 ERA.
#2 – 44.69% of voters picked a 3.61 ERA as the expectations for a #2 starter. 74.07% had a #2 starter as either a 3.30 and a 3.61 ERA.
#3 – 58.15% of voters picked a 4.15 ERA as the expectations for a #3 starter. 87.86% had a #3 starter as either a 3.73 and a 4.15 ERA.
#4 – 40.06% of voters picked a 4.40 ERA as the expectations for a #4 starter. 70.19% had a #4 starter as either a 4.40 and a 4.60 ERA.
#5 – 44.16% of voters picked a 4.75 ERA as the expectations for a #5 starter. 73.71% had a #5 starter as either a 4.75 and a 5.00 ERA.
Those results are drastically different than what Sackmann came up with. However, before I even ran the poll, I did some research on the 2010 results, as I felt the 2006 results were outdated. The results from yesterday aren’t totally unreasonable, despite the big differences between those results and Sackmann’s numbers. The reason for that is that the pitching numbers were better in 2010 compared to 2006. Some of that could be due to the amount of young, talented starters that have emerged the last few years, although I think a bigger role comes with the end of the steroid era.
I studied the 2010 rotations from every team, using the same approach that Sackmann used above. The only difference was that I gave 33 starts to the #1-2 spots, since 32 starts for each spot equals 160 games. Here are the results, by rotation spot:
MLB Average: 3.10
MLB Median: 3.09
AL Average: 3.24
NL Average: 2.99
MLB Best/Worst: 2.08 / 4.17
Average of the Top 15 Rotations: 2.84
Average of the Bottom 15 Rotations: 3.37
MLB Range of the 11-20 Ranked Rotations: 2.88 – 3.35
The results from the number one spot are much higher than they were in 2006. That shouldn’t be a surprise, as 2010 was a great year for pitchers, with 18 qualified starters coming in with an ERA lower than the 3.10 league average. The poll results from yesterday picked this as the most popular choice (and looking at these results, you can see where I got the choices from), although a big majority had a 3.10 ERA or higher as the expectations.
MLB Average: 3.61
MLB Median: 3.58
AL Average: 3.71
NL Average: 3.53
MLB Best/Worst: 2.71 / 4.69
Average of the Top 15 Rotations: 3.31
Average of the Bottom 15 Rotations: 3.91
MLB Range of the 11-20 Ranked Rotations: 3.42 – 3.72
The number two spot looks very similar to the number one spot from the 2006 results. Prior to this research, when I thought of a #2 starter, I was thinking in the 3.90 range, which is a little better than what Sackmann had in his 2006 research. The poll results also nailed this one, although just like the number one starter, the majority of the votes had a 3.61 ERA or higher.
MLB Average: 4.15
MLB Median: 4.12
AL Average: 4.28
NL Average: 4.03
MLB Best/Worst: 3.22 / 5.19
Average of the Top 15 Rotations: 3.73
Average of the Bottom 15 Rotations: 4.56
MLB Range of the 11-20 Ranked Rotations: 3.86 – 4.42
Just like the number two starter results, the number three starter looks similar to the #2 starters in 2006. It’s almost like a better pitcher was added to each rotation, and everyone else moved down a spot. Once again, the poll results picked this result, and once again, the majority of the expectations were better than a 4.15 ERA.
MLB Average: 4.62
MLB Median: 4.57
AL Average: 4.76
NL Average: 4.49
MLB Best/Worst: 3.51 / 6.05
Average of the Top 15 Rotations: 4.20
Average of the Bottom 15 Rotations: 5.03
MLB Range of the 11-20 Ranked Rotations: 4.38 – 4.78
This is where things get interesting. The majority of votes had a 4.40 ERA as the expectations for this spot. Also, 97% of voters picked a 4.60 ERA or better, with two thirds of the overall voters picking better than a 4.40 ERA. Again, this is similar to the 2006 results for the number three spot in the rotation, continuing the trend of the expectations moving down one spot.
MLB Average: 5.69
MLB Median: 5.69
AL Average: 5.68
NL Average: 5.70
MLB Best/Worst: 4.56 / 7.16
Average of the Top 15 Rotations: 5.41
Average of the Bottom 15 Rotations: 5.97
MLB Range of the 11-20 Ranked Rotations: 5.37 – 5.96
The results here are where the expectations get too high. The popular vote had the expectations for a #5 starter at a 4.75 ERA. Almost 90% of the votes had a #5 starter with a 5.00 ERA or better. The average of the top 15 rotations was a 5.41 ERA. Only four teams in the majors had better than a 5.00 ERA: Oakland (4.56), San Francisco (4.70), the Cubs (4.71), and the Phillies (4.83). That is similar to the results from the 2006 poll, although overall the number five starters improved, just like every other rotation spot has over the last four years.
Sackmann also had dividing lines in his research, where he basically took the mid-point between each rotation spot to give a better idea of the range for each rotation spot. This can also be accomplished by getting the average of all of the pitchers for each spot (for example, the mid-point for the #1/2 starters is the average of all of the #1 and #2 pitchers). The dividing lines for 2010:
#1/#2 – 3.36 ERA
#2/#3 – 3.88 ERA
#3/#4 – 4.38 ERA
#4/#5 – 5.15 ERA
That breaks down to the following ranges:
#1 – 3.36 ERA or better
#2 – 3.36 – 3.88 ERA
#3 – 3.88 – 4.38 ERA
#4 – 4.38 – 5.15 ERA
#5 – 5.15 ERA or worse
This research explains a few common phrases. For example, Kevin Correia has been described as a #2 starter on the Pirates, and a #4 starter on a contender. Looking at the information above, the average ERA for a #2 starter on a bottom 15 rotation is 3.91. The average ERA for a #4 starter on a top 15 rotation is 4.20. Correia had a 3.91 ERA in 2009, and in each of the last two years he has been around a 4.20 xFIP.
It also helps explain what people mean when they say a player has the upside of a #2-3 starter, such as my estimate with Owens. The tendency when these evaluations are thrown out are to take the highest number and run with it. I’ll say Owens has the upside of a #2-3 starter, and most people will look at the above and think I’m saying “Owens could have a 3.36 ERA”. What I’m saying is that Owens could be in the 3.36 – 4.38 ERA range, which is the range of the #2-3 starters. Specifically with Owens, I don’t think he will be as high as a 3.36 ERA, but I do think he could break a 4.00 ERA, and I definitely think he’ll be better than a 4.38 ERA.
Finally, we often see comments that the Pirates have nothing but potential #4-5 starters in the farm system, outside of guys like Jameson Taillon who have top of the rotation upside. Based on the poll results, people feel that a #4 starter is a 4.40 ERA pitcher, while a #5 starter is a 4.75 ERA pitcher. Chances are when you see a discussion about a player, and someone says he’s a #4-5 starter, they’re overestimating how good a #4-5 starter should be, and actually referring to a #3-4 starter.
Tomorrow I’ll take a look at the Pirates and their rotation as it currently sits, as well as some of the prospects who could arrive in 2011, to see what exactly the Pirates have to work with in 2011.