A Bullpen Philosophy


I don’t blame Octavio Dotel for yesterday’s implosion (well, actually, I do – but I don’t blame him alone)… I think a healthy slice of that blame should fall to Jim Leyland.

Stat-heads tend to assume that everything that doesn’t stay the same from night to night as genuinely random – in a (duh) statistical sense. That would mean that a the fact that a pitcher got hit hard the night before doesn’t mean he’s likelier to get hit hard tonight and that each and every pitch is equally likely to find it’s spot in the zone, no matter where the previous pitch wound up. I think about this game in what is probably an overly statistical way, but even I don’t buy that. Sometimes guys have it. Sometimes they don’t. It’s hard to tell why they didn’t have it, and they might very well have it tomorrow, but tonight if a guy just doesn’t have a feel for the ball he’s a tremendous liability.

It has long been my personal opinion that Jim Leyland and many other managers are simply too slow to pull a reliever who doesn’t have it on any given night. A starter might be able to stay in the game long enough after a rough first inning to start to get a feel for the ball and the strike zone, a reliever only gets that first inning. I propose a simple rule – since Leyland seems to like rules just like roles: If a guy walks the first batter he faces – he’s out. Perhaps some flexibility might be preserved for two strike walks where the hitter battles off pitch after pitch. That’s not all on the reliever.

Surely there have been plenty of cases where a reliever pulled himself together after a leadoff walk – but I’d bet that those situations tend to end badly. I’d be perfectly willing to sacrifice the occasional leadoff walk followed by three quick outs to avoid ever having to sit through a leadoff walk followed by another walk followed by two wild pitches. AND LEYLAND DIDN’T EVEN TAKE HIM OUT AFTER THAT! When Dotel entered Monday’s game, the Mariners had an 8% chance to win the game. That’s it. After walking Ryan on 5 pitches, Dotel had sent a strong signal that he wasn’t ‘on’ that night – but the Mariners still only had a 17% chance to win. After he walked Suzuki on 4 pitches it should have been obvious – but the Mariners only had a 31% chance to win. By the time Leyland did take the ball out of Dotel’s hand, the score was tied and the Mariners had upped their chances to 80%. I count that as 9% of a loss on Dotel and 63% of a loss on Leyland (if you prefer my rule) or 23% and 49% (if you prefer the rule any sane manager should be following – to pull a guy who misses the zone in 8 of his first 9 pitches).

Adopting this sort of a bullpen scheme would – of course – have an impact on how and when you warm up your relievers. There would have to be somebody throwing whenever anybody entered the game, or else you won’t have anyone to bring in when you need them. I’m not sure that it would do much to diminish the effectiveness of a ‘pen, though, a ‘change’ isn’t really synonymous with a ‘cost’. Let us know what you think.