John Mclaren will cost the Mariners this year

This post is about a team that Joel loves, the Seattle Mariners. That is its only connection to Joel's 21st birthday.

If last night's Mariners game was a season preview, we are in for a long, frustrating one. The Mariners attempted two steals, and got caught twice. This is a poor strategy.

Here is a run expectancy matrix: http://www.tangotiger.net/RE9902.html. With a runner on and no outs, a team can expect to score, over the course of a season, about 0.95 runs an inning. With a runner on second and no outs (a successful steal), the team can expect 1.18 runs, or an increase of .23 runs. With no on and one out (a caught stealing) a team can expect 0.29 runs, or a loss of .66 runs. So a caught stealing is about three times worse than a stolen base is better, both compared with staying pat. That is, a team needs a 75 percent success rate just to break even. (This discussion could, of course, be extended to every other out/runner situation, but a rough argument here can stand in for a more refined one). Last year, the Mariners stole 80 bases, and were caught 31 times, for a success rate of 72 percent. But with Mclaren threatening to run more, his only option is to run more with players who are not as good at stealing, so the success percentage is sure to go down. Given that the Mariners are already below the 75 percent level needed just to break even, this year will only get worse.

There are two quick retorts that merit discussion. First, Jose Lopez was not told to attempt the steal on third. This would work well as an argument if Mclaren had spent spring training talking about run expectancies and the virtue of not running, but he did not. He told his team, his coaches, and the media that he wanted the team to run a lot. Lopez was just applying specifically what his manager pounded into him generally. Second, a team like the Mariners has to run, because they will not hit many home runs. This strikes me as an even stronger argument not to run. The Mariners only get 27 outs a game, and fewer of those opportunities will result in guaranteed scores via home runs. Why give outs away when the return is tiny compared to the risk?

There is a third retort, which is that announcers always applaud when a manager tries to steal. But being a good manager means doing what will give you the best chance of winning games, not what Joe Morgan thinks is great. If we continue to run a lot, we will lose more games than we would otherwise.