Recently here at Rotographs, we released our rest of season rankings for starting pitchers, I had Tim Lincecum rated as the sixth-highest fantasy pitcher. To say the least, I took a little heat for that ranking. I will try to justify the ranking today.
To begin with, here is a comparison between two pitchers from the beginning of the season to the All-Star break.
Pitcher A: 5.66 ERA, 0.347 BABIP, 11.7 K/9, 1.8 BB/9
Pitcher B: 6.08 ERA, 0.335 BABIP, 9.8 K/9, 4.7 BB/9
Both of these pitchers put up a high ERA. They both struck out a ton of batters. Also, they were getting hit around quite a bit as seen by the high BABIPs. The one difference is that Pitcher B walked a few more batters. Pitcher A is Zack Greinke in 2011, while pitcher B is obviously Tim Lincecum this year.
After the horrible start last season, Zack began getting the results in accordance with his peripheral stats. Over the second half of the season, he had 9 wins, a 2.59 ERA, 1.16 WHIP and put up those numbers while his K/9 dropped to 9.4. He is an example of a pitcher who was putting up great peripheral stats, like Lincecum, but just not getting good results. An example of one other pitcher is not how I based my decision on ranking Tim so high, but it shows that pitchers can turn their season around.
I came up with my rankings using a formula I created last fall to rate fantasy pitchers. When I created the formula, I was just wanting to take most of the noise out of determining how to value SP. After looking at several different formulas, I ended up using one that only looked at the K% and BB% for the pitchers. By using percentage values, some levels of pitcher BABIP is taken into account. With the formula, Tim’s 2012 value (no IP used) was 0.159. It was the ninth-highest value of the starting pitchers when I ran the my rankings during the All-Star Break. Some pitchers ranked ahead of him had possible inning limits placed on them, like Chris Sale and Stephen Strasburg, so he moved up to sixth place.
At that point, he may not of seemed to be the sixth best pitcher with a:
Those stats, besides the strikeouts, are horrible. To put their crappiness into perspective, I ranked him with other starters with comparable IP. Here at FanGraphs, we don’t have a way to create leader boards using the All-Star game as a splitting point, so I couldn’t go back and create a ranking of fantasy stats for all pitchers at that point in the season. I had to just use the season stats to date. Taking all the current stats and using Zack Sanders’ method of ranking SP, Tim is the 84th-ranked starting pitcher (of 91 pitchers with at least 100 IP). Those values include his last two starts which were good, so he probably had a lower ranking at the halfway point. Even though he has the 13th-highest strikeout total, he is producing like Kevin Millwood or Rick Porcello.
I can understand the confusion of ranking a pitcher 6th best when he was producing more like the sixth-worst pitcher in the league.
The problem with projecting Timmy is that his level of suckitude, especially when considering his ERA, is unprecedented for a pitcher putting up his K and BB numbers. Looking back over last 10 years of stats for pitchers in the first 4 months of the season, not one pitcher has a Pitcher Valuation Ranking as high as Tim with a comparable ERA. It is not even close. Using pitchers from the last 10 years who had:
1. Pitched at least 80 innings from the start of the season to the end of July
2. Posted an ERA within 0.5 Runs of Lincecum.
None were “better” than him using my pitcher valuation. Here are the top 5:
Rank: Name, Year, Pitcher Valuation, ERA
1. Tim Lincecum, 2012, 0.159, 5.72
2. Carl Pavano, 2009, 0.132, 5.68
3. Bronson Arroyo, 2008, 0.131,5.79
4. Jeremy Bonderman, 2004, 0.125, 6.05
5. Francisco Liriano, 2009, 0.123, 5.60
Looking at it another way, here are pitchers with similar Pitcher Valuations (K and BB are within 1% also) and their corresponding ERA:
Name, Season, Pitcher Valuation, ERA
Tim Lincecum, 2012, 0.159, 5.72
Bud Norris, 2003, 0.161, 3.39
Yovani Gallardo, 2012, 0.158, 3.76
A.J. Burnett, 2005, 0.158, 3.32
Edinson Volquez, 2008, 0.155, 2.74
By having such an extreme difference in talent and results, it is tough to get any good examples of how similar pitchers have performed over the rest of the season. I took 20 pitchers’ with the best Valuation Values and had a similar ERA to Tim this year. With that set of players, I looked to see how they performed from August to the season’s end. Their ROS ERA was, on average, lower by a value of 1.08. Five of the 20 pitchers had an ERA 2 or more runs lower while 3 actually had an ERA greater than their previous ERA.
Even though these pitchers’ K and BB numbers were worse than Lincecum’s, I expected to see more of a change in their ERA. Looking at the numbers a little further, even more regression was at work besides the ERA value. These pitchers saw there Valuation numbers drop in the last two months from 0.112 to 0.092 (K% went from 17.8% to 17.0% and BB% went from 8.4% to 9.8%). Regression was at work with the ERA and with K and BB numbers.
I am not 100% sure how to measure Lincecum’s ROS talent level. Generally, the rest of the pitchers I ranked lined up with the rest of the rankings. I have thought of putting some kind of BABIP and/or LOB value into my formula, but when I originally created the formula they added little if any value. One item I am considering looking into are K% and BB% numbers with runners on base and the bases empty since K and BB rates stabilize quickly. Tim has a distinctive split in 2012 with a 4.1 K/BB value with the bases empty and a 1.2 K/BB value with runners on base. These values are significantly different than his career 3.0 K/BB with the base empty and 2.8 K/BB with runners on base.
Has Tim Lincecum performed like the 6th best pitcher in baseball? Not even close. There is data that points to him being the 6th worst in the league. I will stand by my ranking because, over time, pitchers eventually begin to perform closer to their peripheral stats. Since I released the rankings, Tim as done a good job of making me look good by winning 1 of this starts and posting a 1.20 ERA and 0.867 WHIP. Two games aren’t going to show that he is back, but they help. Sometimes it takes a while for a pitcher to produce according to their strikeout and walk numbers. This season, it is taking Tim a historically long time to start to produce to that level.