Projecting RBIs

I hate RBIs. I hate what they stand for. I hate when announcers talk about how great RBIs are. I hate when “Hank Aaron’s greatest record is the RBI record.” I hate when players are picked in fantasy baseball leagues because they “drive guys in.” I hate when Joe Morgan writes books about baseball and says that, in fantasy, you need an “RBI man.” 

But because the majority of us are forced to use RBI as a category in our fantasy leagues we need a way to reliably factor them into our player valuations. One way is to just assume that better players will get more RBIs. But this is not the case. We could also just look at team offense, but again this doesn’t factor in how often a player will convert RBI opportunities. 

In my opinion the best and only way to semi-reliably project RBIs is to look at batting position, team offense, and player ability. The first formula I will look at is pretty simple. For #3 hitter s in the AL Projected RBI = OPS+(% relative to other #3 hitters)*100*OPS+(of team). Lets run that down. OPS+ relative to other #3 batters will show you how much better said player is from other #3 batters. The average third hitter has an OPS of .805 in the American League. Mark Teixeira had an OPS of .963 last year. That means his OPSbo+ would be 120%. That makes him 20% better than the average third hitter. The 100 is the average number of runs batted in by #3 hitters in the league. If we do the first half of the formula we get 120 RBI. In an average American League lineup, Mark Teixeira should drive in 120 runs. 

But the Yankees aren’t an average American League lineup. We need to find out how money runs he will drive in on the Yankees. So lets look at the Yankees OPS+, which happens to be 102%. If we multiply Mark Teixeira’s neutral RBI by the amount the Yankees are better than the neutral team we should get his RBIs right? That gives him 122.4 runs batted in. Seems reasonable. Now we do have one problem here. We don’t know if we are correctly factoring in any of this. Maybe (actually probably) RBIs have far more to do with who bats in front of you than total team offense. Though by using batting order averages we factor this in a bit, we don’t have the numbers to back this up. So lets take a look at Bobby Abreu in 2008.

Abreu hit #3 for the Yankees, who we already know is a 102% team. If we multiply 100 by 1.02 we get 102. So the average #3 hitter should drive in 102 runs on the Yankees. Bobby Abreu’s OPS is .842. The league average #3 hitter has a OPS of .805. That means Abreu has an OPSbo+ of 106%. 102 times 1.06 is equal to 108. Abreu drove in a 112 runs last year. Now this isn’t dead on, but there is certainly a lot of random luck and chance involved. It seems as though we might be on to something. But of course, there isn’t much proof here. Much more research needs to be put into here before we can come to a conclusion. 

I think something that needs to be stressed is that we are operating under the assumption that batting order position is of much greater importance than any other factor in RBI totals. While this may not be the case, we run into all sorts of problems if we start looking at player ability as paramount. The problem is that a players RBI total is effected by countless outside forces. When we look at a batting spots RBI total, we can take away the “who put up these numbers?” factor by just looking at the average, and then finding out how much better our target player is than the average. If we look at Abreu, we just can’t take away the team offense and batting order factors, as much as we may try. In my opinion RBI are probably the last thing you want to calculate. They don’t effect anything else, and everything else effects them. 

I don’t think we should draw a massive conclusion from this other than that RBI totals are effected a lot by luck and random chance. We know that team offense, batting position, and player ability are factors however I’m not sure we are weighing them correctly. As far as I can tell, this projection system does a good job of approximating how many RBI a player should put up, however there is a lot more time that could be spent on this topic.