Everyone who handicaps baseball pays close attention to the starting pitchers. That only makes sense. Far fewer people pay more than passing attention to the bullpens. That makes far less sense. Starters typically are only good for about six innings, so the game is going to be in the hands of the bullpen when the game is on the line. I'm not going to go to the effort of proving this, but I would guess that the bullpen is responsible for more than a third of outcomes even if they are typically only in a third of each game, on average.

There's good news in all this. Because most people aren't paying as much attention to bullpens as they probably deserve, you can gain an edge over most people if you are on top of the bullpens. Of course, you aren't going to be able to do it perfectly. We know which pitchers will be starting, but we can only guess which relievers will be called upon, and that can change significantly based on how the game plays out. Still, by tracking bullpens you can find general areas where you have an advantage that you can try to exploit. It's like card counting - if a deck is highly advantageous you aren't going to win every hand. It just means that you are more likely to win. Similarly, you can find cases where a bullpen is more likely to succeed or fail than normal, and you can bet accordingly.

There are many different ways to look at bullpens and track their status. As with many things in handicapping, it is probably more important that you do something to track them than that you do it perfectly. As you are thinking about bullpens, here are three different approaches you might want to keep in mind:

Short term - Even if a bullpen is generally well staffed and effective they can get themselves into trouble if their starters have had a bad week. Even just two bad outings in a row from the starting staff can cause the relievers to throw a lot of innings and leave them less fresh and prepared than they would otherwise be. A bad outing back-to-back with an extra innings game can have the same effect. Basically, you are just looking for situations where the manager is going to have less bullpen options than he would normally have, and where those options might not be as fresh as they could be even if they are available. This could cause managers to leave their starter in for longer than they should to try to alleviate the pressure on their pen. The situation can be especially problematic for managers if a rough night game by a bullpen is followed by an afternoon game the next day, or if the team has ugly travel following a couple of rough games. Anywhere you can find a situation where a bullpen is going to be sub-par is your goal.

There is another thing to consider here. This is especially powerful if you have a team that has a statistically strong bullpen that is in rough shape. A lot of people just look at the overall bullpen stats to evaluate the relievers, so you have a powerful edge if you find a situation in which the current expectations differ significantly from what the bullpen is capable of at their best.

Longer term - If a reliever has been in the bullpen for a few years then it's easy to look back and see how many innings he typically pitches in a season. You can then look at how many innings he has pitched so far to find situations where a pitcher is being overused. This can happen because of injuries or shortages in staff. An overused reliever is quite likely to lose effectiveness as the season progresses. Even if a pitcher doesn't have a bullpen track record you can approximate. I use 90 innings as a benchmark unless I have reason to believe something else. A pitcher that is on pace for significantly more than 90 innings is likely heading towards troubles, and I will be on the lookout for ways to take advantage of that down the line. This is especially advantageous when a team is having troubles with their rotation - more than one reliever could be heading towards burnout, and that could create serious bullpen problems down the road.

Tendencies - Before you get too excited about any particular bullpen situation that you have noticed, it is important that you spend some time getting to know the tendencies of the managers that handle the bullpens. Some managers have a quick hook with their starters, so their bullpens handle more innings. If they have been there for a while then their bullpen staff will be used to the extra work, and will even have been chosen for their ability to handle it. Others don't like to use their bullpens much, so they will be particularly vulnerable if the bullpen has been in action a lot over a short period of time. Being able to compare the way a bullpen has been used recently to the way a manager typically uses a bullpen can be very valuable.