Remember me

Boredom Prevails

June 8, 2012

                Fortunately, I have something in the hopper that I like for you guys tomorrow or the next day, because, in terms of the pitching rankings, nothing continues to happen.   An astonishing amount of nothing.    Justin Verlander has pitched a couple of off games and could be in danger of slipping from his top spot, except that Kershaw, the only guy within hailing distance of him, didn’t pitch real well, either, so Verlander is still in a strong first-place position.    The top 10 remain essentially unchanged in the last month; some changes of position but the same top 10 as a month ago.    This is unusual, that there would be no one making a move within the top 10, but. . .that’s where we are.    I’ll present the data with little comment; blue background indicates a pitcher who has been pitching better than those around him in the last month; red indicates a pitcher who has not been pitching as well as those around him in the rankings:

 

AApitchers1 AApitchers2 AApitchers3 AApitchers4 AApitchers5

 

The Desert

                In a sense, all rankings in this system under 400.00 are phony rankings:  in this sense, that no one is actually a bad enough pitcher to have truly established a performance level under 400.   If you posted a Game Score of 35, 35, 35 every time out, that would be equivalent to a ranking in this system of 350.00—but if you posted a Game Score about 35 every time out, you’d be out of  a job within a month or, if you’re defending a Cy Young Award, two months.    A pitcher has to post Game Scores around 40, on average, just to stay in the rotation.    The Rankings of 300 to 400 are inconsistent with the theory of the ranking system in a certain sense.

                But I had good reasons, I think, to set the system up the way it is.    I could have made 400 the base, started pitchers out at 400, and moved them up from there, up from the replacement level line.   If I had done that Stephen Strasburg would now rank about 45th in the majors, rather than 101st, and I know that a lot of people would like that better.   

                But if I had done that, to begin with, about 60 pitchers who are currently ranked wouldn’t be in the system at all.    It can happen, and occasionally does, that a young pitcher comes up, makes a start or even two starts, but pitches so poorly that he fails to show up in the rankings.   In the last week, for example, Garrett Richards and Casey Crosby have made starts, but don’t show up in the rankings because they didn’t pitch well enough to suggest that they deserve to be ranked.   It does happen.

                But if we started with a base of 400, rather than 300, that would happen a lot; there would be a lot of young pitchers hanging around, trying to get established in the rotation, who weren’t in the rankings.    I’d rather have them in there than not have them in there.

                If we bottomed out the system at 400, we’d have to change the off-season adjustment process, or half the league would start the new season at 400.   I’d rather have a system that distinguishes between those who proved something last year and those who proved nothing.   

                I call the area between 300 and 400 The Desert.    It is the long, dry stretch that a young pitcher must crawl across, must struggle across somehow, to reach the heart of the ranking system, that place where he is listed next to Ivan Nova and J. A. Happ and Jake Westbrook.  

                What we are doing, really, is this:  We are making the young pitcher prove that he can stay healthy and stay in the rotation and continue to produce, before we are willing to regard him as a "real" pitcher.   The reality is, most young pitchers will die in the desert.   Most of them are never going to make it to 400.   

                Remember how well Kyle Drabek was pitching a month ago?  Remember how well Luke Harrell was pitching early in the season, and Jake Arrieta, and remember those strong starts that Danny Duffy had?    Do you remember Mike Minor having that stretch of good games, and Jeanmar Gomez?    Remember Drew Smyly being mentioned as a Cy Young candidate?

                Young pitchers do that; they pitch well three, four, five games, and then they fall apart on you.   I used to call it Sam’s Law, after Sam Reich:   Young pitchers will break your heart.   They show flashes of brilliance, then they disintegrate in front of your eyes.

                I don’t want to rank them among established major league pitchers—even among not-very-good established pitchers, until they get through that period.   We’re far enough into the season now that the new pitchers and the pitchers coming back from injury are almost out of the desert.     Strasburg is at 389.0 now—89% of the way across the desert—and he now actually ranks in the top half of all ranked starting pitchers.   Johan Santana is in the 380s, Yu Darvish just short of 380, Lance Lynn at 377, Felix Doubrant at 372.    The good ones are starting to separate themselves from the herd.   That’s the way I think the system should operate.

 
 

COMMENTS (6 Comments, most recent shown first)

kcbbfan
Thank you for the color coding, I think it makes the chart even more readable.
10:22 AM Jun 9th
 
jemanji
It's awesome that with the "World's #1 Starting Pitcher" system we have a way to capture pitchers that is not focused on the previous season's ERA. Why should last year's ERA (or xFIP or whatever) be the focus? Why not the last 3 months', or 13 months', or whatever?

"World's #1" captures a pitcher's performance over a longer, and less artificial, time frame than do other methods. If there were no other reason for the system, I think that the *timeframe captured* would be reason enough for its place in evaluation ...

.........................

As far as "crawling across the desert," again from chess there is a decades-old wheel that has been invented ... :- )

First question: what's wrong with setting the starting point at 400, and then publishing ratings of players who have performed below the presumptive 40.0/game minimum standard of competence? In chess you do have players rated below a presumed minimum level of competence.

There's a funny little story about Bobby Fischer, in recluse around 1975, meeting a fellow GM at a ladyfriend's house. Fischer, whose chess rating was 2800 (2200 being master, 2400 international player, 2600 top ten, 2700 being the #2), found out that the housewife hostess was a tournament player. He immediately demanded to know her rating. "I'm ashamed to tell you," she said. OUT WITH IT!, Fischer bellowed. 2100? 1900? What?

"No, I'm only 1100," she said. Fischer sat stunned. "Do they go that LOW?", he whispered, genuinely astonished.

Super-low ratings can accurately reflect true skill and at the same time be sources of fascination as one struggling player is compared to another...

..............

As far as requiring players to crawl across the desert ... in chess (for what it's worth) the goal is accurate ratings, the quicker the better. What are the ratings going to be *used* for? To predict performance -- this next game.

Players who have played under 20 games -- here, the "desert" pitchers -- do have ratings. Their ratings are "provisional." Rather than leave them off the ratings chart, they'll have the rating "375/12" next to their names -- meaning that they have averaged 37.5 points per start for 12 starts. After 20 games, the rating is established. During the provisional period, their ratings are volatile: if a guy has scored 50%, his rating is simply equal to the average of his opponents up to that point.

In a chess system, Strasburg's provisional rating after 15 starts might be 610/15. If it's really that high, good for him - that has been his actual performance. If his rating becomes "established" at 610 after 20 starts, and that's too high, he'll quickly get embarrassed as his opponents outmatch him and his rating tumbles. It doesn't happen often.

If this system has the de facto #12 pitcher in the world ranked #60 for a while, that has its advantages too. Last year, Michael Pineda looked like a top-20 guy. There's something to be said for doing it over a period of years...




4:39 AM Jun 9th
 
bjames
Well. . .it's hard to stay in the rotation at 400. If you averaged a Game Score of 39.0, you WOULD drop out of the rotation in something less than a year, absolutely, but at 40-42, you'd be in constant danger of dropping out of the rotation. The cluster of injured pitchers around 390 is just a coincidence, and I still don't really understand what you're trying to do with it.
4:06 PM Jun 8th
 
Trailbzr
Using the values in my example: If at the end of the season, the pitchers with scores about 400 all have recent starts next to their names, then that's not the level at which they get replaced. But if most of the pitchers at 380 haven't started recently, then something between 380 and 400 represents the line below which they'll get dropped from the rotation.

2:07 PM Jun 8th
 
bjames
I don't understand what you're asking. What is "essentially a numerical value of replacement level?"​
1:59 PM Jun 8th
 
Trailbzr
Then will this finish the year with essentially a numerical value of replacement level? There's a bunch of guys around 380-90 who haven't started since the chart began. So are we likely to finish with a bunch of active 400s; and a lot of blank space to the left of scores slightly less?

11:19 AM Jun 8th
 
 
©2024 Be Jolly, Inc. All Rights Reserved.|Powered by Sports Info Solutions|Terms & Conditions|Privacy Policy