Blue Jay Way II: A Real Gamble

I don't want to be mistaken for an old baseball fuddy-duddy. Last year I complained about analytics, but I did so as a fellow traveler. I was an early devotee of Bill James and still feel many of the points he made represented an advance in baseball thinking. I've continued to read him on and off over the years and believe there is an important role for analytics in baseball. If I can reframe my concern, the problem is not that it has "gone too far." The problem is that it is being misapplied. Analytics refers to the use of advanced statistical analysis to make decisions in baseball. Baseball managers have been using stats for a long time (think Earl Weaver and his recipe cards), as have players. The problem occurs when statistical evidence is misapplied. I'll give you an example. 

Generally, most managers today carry large bullpens (a minimum of 7, sometimes 8) relievers. The idea is to bring in a reliever at the start of an inning so that the bases are "clean" and to have the reliever pitch no more than one inning. And, this is true. If you limit any given pitcher's exposure while creating as favourable as possible situation in which they pitch ... that pitcher will do better. It does not, however, mean (1) that they are a better pitcher than they were. It just means that stats are disguising their ability because they've been pitching in favourable conditions, and (2) it doesn't necessarily save runs because someone had to pitch in an unfavourable situation and that is usually the previous pitcher. Imagine this situation, a starter labours into the fifth where there are runnings on second and third with one out. What should you do? If you bring in a reliever, you are asking the reliever to pitch in an unfavourable situation. If you leave in the starter, they are pitching in an unfavourable situation, And, you are running the risk of leaving that pitcher in one batter too long. Anecdotally, this seemed to be one of the Jays problems. In order to create a favourable situation for the bullpen, they seemed to be leaving the starter in -- particularly near the end of the season when they needed wins -- one batter too long. And, because they burned through so many pitchers, they often turned to the wrong pitcher in the pen to get them out of trouble.  Sometimes, those pitchers just were not up to the job. 

You see the situation: analytics is not solving the problem. It is redistributing the problem. Said differently, I am not certain -- as I posted last year -- that analytics is doing much to help the Jays win.  So, what should the Jays do this year? In my last post, I intimated that the Jays front office gambled last year. They downgraded the offense to bring in potentially short-term solutions to their defensive problems. The problem with that gamble was that: (1) it did not work, and (2) it created the need for more personal gambles down the road (aka, this off-season). 

The Jays do need to address the holes in their lineup, but how they do it depends on the nature of the gamble that they want to take. For instance, for me, I'd like to find out if Schneider really can play second base. That's a gamble because one is gambling on player development. A safer solution would be the cast of the thousands we saw last year which is actually a way of avoiding a decision while hoping someone catches fire (last year, it turned out to be Schneider but too late in the day to substantively affect the kind of team the Jays were). 

A different kind of gamble might be to make use of their team in a different kind of way. I don't know if that is possible. For instance, I'd love to see starters go deeper into games, lessening the need to carry so many arms in the bullpen and making more space on the bench for platoon players, defensive subs, pinch runners, and the like. In other words, rather than trying to make analytics work like everyone else in the league -- something I don't think we have any reason to believe the Jays can do -- I'd like to see a real gamble that involves doing something different. Tampa Bay, a team that makes heavy use of analytics, has also been a team that innovated. Analytics was, in fact, an innovation at one point in time, as was an opener, as was the one-inning bullpen pitcher, as was the supersub. TB has stayed in the hunt because they have confidence in their own ability to judge talent (they don't need to turn to free agency to find good players), because they have tended to have really good player development, and because they have not been afraid to take chances. This may be the result of Tampa being Tampa. They might have to take chances because their small market does not allow them to do anything else.  

The Jays have been oddly the opposite. They have turned to free agency because they have not developed the talent they need. They have been indecisive and cost players (like Biggio) the kind of playing time they needed to develop. And, they have not taken chances but rather followed other's lead (their organization of the pitching staff mirrored Tampa's, for instance).  Will the Jays break from their established routes? I don't think they will this year, but another subpar season and they may have no choice. And, for my buck, that would not be a horrible thing.