Back in 1984 when we started this wild ride, David and I religiously followed the rules of the original Rotisserie book which talked about keeping stats over an entire season to determine a champion.
The categories that were used originally were split evenly between offense and pitching. On the offensive side we tracked: Home Runs, Runs Batted In, Batting Average and Stolen Bases. On the pitching side of things we used: Wins, Saves, Earned Run Average and WHiP (which is a calculation of Walks + Hits divided by Innings Pitched).
Those categories served us well for many years, until Matt Bentel, The Idea Man, pushed us to think outside the box and consider new categories. It's possible that the idea of changing categories had been broached prior to 2001 but Matt provided the consistent voice on why new categories were necessary.
Initially there was decent resistance, which makes sense since we had been using the same original categories for seventeen years. But when you look at the original categories, there are some major flaws.
Home Runs - Puts a lot of emphasis on the big bopper, but has absolutely no reward for a guy who hits a lot of doubles and triples. Also can favor a marginal hitter who plays in a homer friendly ballpark.
Batting Average - Great if you have Tony Gwynn, but sucks if you have Adam Dunn. Dunn will get on base 38% of the time (or at least did in his prime), but the majority of that is from walks. His .214 batting average can kill a team.
Wins - There were a lot of good pitchers that we avoided like the plague in our early years. If a good pitcher played for a terrible team, there was a good chance the pitcher could have a decent ERA and WHiP, but have hardly any wins. If the pitcher gave up one run in eight innings but received no run support, you would walk way without a win even though your guy pitched great. Plus, the category was, at best, a zero sum gain. Two pitchers couldn't receive a win in one game, even if they both pitched a complete game giving up one and no runs respectively. So the drafting strategy was to try and pick a guy who could work himself into a win based on how the team behind him was playing.
Saves - Fine, we fell in love with relievers. But the guys doing the heavy lifting in the seventh and eighth innings were meaningless to us unless they picked up cheap wins or filled in for the closer due to injury or ineffectiveness.
Matt's prodding and our ensuing discussions made us realize there were better options. But what do we pick? We wanted something that would challenge us as team owners to put together the best team possible based on each player's talent and statistics. We didn't want to avoid Pittsburgh pitchers simply because the team was going to win only 60 times all year. We also strived to find categories that tapped in to the value of virtually every talented ballplayer. We didn't want only home run hitters or ace starting pitchers.
In addition to those considerations we needed to use categories that were somewhat easy to understand and research. There was still the element of excitement of being able to look at a box score and have an idea how your team did that was important to us. Sure we could Bill James the hell out of our categories, but then we wouldn't know how we were doing until a computer capabale of keeping NASA happy could crunch our numbers and tell us how good our team was.
So where do we start? Well, we did a lot of league discussions on our forum page. We used common sense (or as close to common sense as our owners could muster). We looked at what other leagues were doing. We researched, we argued and, ultimately, we voted.
What we came up with has served the CFCL for the last thirteen years. We still broke things evenly between pitching and offense. We went from eight categories to ten. Here's what we chose, or stayed with, and why.
Offense:
On Base Percentage - A better reflection of the success of what a batter is ultimately supposed to do - get on base.
Total Bases - Allows us to understand and benefit from the value of, not only the homerun hitter, but the guy who hits the gaps consistently.
Runs Batted In - Sabermetrics will poo-poo the value of RBIs since it largely is determined by where a player bats in the lineup and who bats ahead of him. That being said, this is one of those instances where we didn't want to get rid of a category we had all grown up with.
Runs - Previously this category wasn't used. Now we could include players at the top of the lineup who didn't bat in a lot of runs, but got on base and, importantly, scored the runs to help their team win.
Stolen Bases - We didn't change this category. Some leagues will use the net of stolen bases minus caught stealing which is pretty slick. It puts extra value on the player who will steal effectively and not run into a lot of outs. I think on this one we started to get concerned about having to do too much math to determine our standings which could diminish a big component of being in the league in the first place - FUN.
Pitching:
ERA - No need to change this, although it is influenced a bit by the defense behind a pitcher and the park they are pitching in.
WHiP - Again, no need to change this.
Quality Starts - Now we get to value a pitcher for what they have done, rather than what their team did behind them (run support) in a given game.
K/BB - This is strikeout divided by walks. We used this rather than simply strikeouts because if you have a pitcher with ten strikeouts, that's pretty good. But if he also gave up eight walks in the same game - first of all he must be freakin' tired. But he also didn't put his team in the best position to win. And his ten strikeouts would then be more valuable than a guy who struckout four but only walked one.
Holds+Saves - Now we can value every worthwhile pitcher in the bullpen, not just the 9th inning guy.
Category changes could easily be higher on the list of significant changes . . . until you see the top three. Category changes are the fourth most significant change in CFCL history.
No comments:
Post a Comment