A pitcher’s record is not meaningful

Adrian sent me a link today to a list of pitchers who had recorded a win without facing a single batter. Seeing this list reminded me of my total disdain for the practice of assigning wins and losses to pitchers in baseball. A pitcher’s record is not a meaningful measure of a his skill and, as far as I’m concerned, doesn’t really have any meaning at all.

First off, it just doesn’t make any sense to credit a pitcher with a win. A pitcher, through his efforts on the mound, cannot win a game. He can prevent runs from being scored, thus preventing a loss, but there is nothing he can do to actually win a game (excluding potential offensive contributions, which are not explicity counted in his win-loss record). He could throw a perfect game for 15 innings, but if the offensive side of his team does nothing, he’s not winning the game.

Second, a pitcher can pitch a great game and not get a win through no fault of his own. This happens all the time. Imagine two different starters on different teams, not playing each other. They both pitch 7 innings, allowing 3 hits each but no runs. Before they leave the game, their teams score at least one run to take the lead. Starter A then leaves the game and his bullpen blows the game for him. Starter B leaves the game and his bullpen holds on for the win. Starter A gets a no-decision and starter B gets the win even though they had identical performances. That doesn’t make sense.

Third, a pitcher can turn in a terrible performance and still get a win. This happens less frequently, but is by no means uncommon. This time, let’s say that starter C plays for a team that generally scores lots of runs and starter D plays for a team that hardly scores any runs. Starter C goes 5 innings and gives up 8 runs but his offense scores 10 runs before he leaves the game. His bullpen maintains the lead through the end of the game and he gets the win. On the other hand, starter D goes 8 innings, giving up only 1 run on 4 hits, but his offense doesn’t manage to score any runs during the entire game and he is tagged with the loss. In this case, it’s obvious that starter D turned in the better performance, yet he got the loss and starter C got the win.

Fourth, which pitcher gets the win depends on when the offense scores runs. This means that if starter B from above had pitched 7 scoreless innings but his offense didn’t score any runs until, say, the 9th, then he wouldn’t get the win; rather the guy who had pitched in the half-inning prior to the runs being scored would get the win. This also doesn’t make any sense.

Of course, my favorite illustration of the silliness of the way wins are assigned is the case where a team’s closer comes in and blows the save in the top of the 9th only to have his offense score more runs in the bottom of the 9th, netting him the win. That’s rich.

In baseball, stats are constantly used to compare players and evaluate their skill relative to one another. Thus, it is important that those stats are accurate measures of their skill. Using wins in this capacity is not meaningful because, as we have just seen, there are innumerable ways that a good performance could not result in a pitcher getting a win or that a bad performance could still get a pitcher a win. Thus, I think it should be clear that wins are not a good indicator of skill and therefore not a good basis for comparing one player to another. In addition, because wins are so tied to the performance of a team as a whole, wins are a poor indicator of how well a pitcher might perform if he were traded to a different team.

In the end, the main problem here is that baseball is a team game. If a pitcher can strike out every batter he sees, then you can commend him for an outstanding individual performance. However, once a ball collides with a bat, the rest of the team is responsible for what happens. So, in my mind, it doesn’t make sense to pin a win or loss on a pitcher. What we need to do is assign a win or loss to the whole team. Oh wait, we already do that.

2 Responses to “A pitcher’s record is not meaningful”

  1. August 2, 2007 at 9:30 pm

    First, thanks for the link to my blog post on B-R.com’s Stat of the Day. Second, I totally agree with your analysis. This is why W/L records must always be taken with a grain of salt, and why so many other statistics are always considered. If you want to get truly analytical about it, going to a system like Bill James’ Win Shares is really much better.
    Without a doubt, the pitcher most hurt by the screwiness of wins and losses is Bert Blyleven. He lost many, many games 1-0 and 2-1, and had a lot of no-decisions in there too, and this is the only reason he’s not in the Hall of Fame. I once saw a report that said if he had had just average luck, he would have won something like 20-25 more games, and he’d have gotten into the HOF right away. I will try to make a post about this on the blog coming up soon.

  2. 2 Colin
    August 2, 2007 at 9:52 pm

    Thanks for the comment, Andy.

    I agree that win shares is a better way of assigning credit.

    My whole aversion to wins came last year when the people in my fantasy league were trying to decide on which stats to use for scoring. I wanted to use stats that reflected only a player’s individual performance and not, in some way, that of his team. Since wins is so strongly coupled to the rest of the team, I strenuously objected, but I was overruled.

    Anyway, I would be interested in doing some more quantitative analysis, but I don’t really have time for that right now.

    Bert Blyleven definitely lost out not only because of the screwiness of how wins and losses are assigned, but because people seem to view them as meaningful, which I believe is a mistake. I hope he makes it into the Hall. I’m sure he will eventually.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


August 2007

Recent Twitterings

Follow Me on Twitter

RSS That to which I am listening

  • An error has occurred; the feed is probably down. Try again later.

%d bloggers like this: