You don't have to have the same talent. The inspiration for starting this thread was the publication of PECOTA projections for the 2006 season. The idea was to use the objective, quantifiable statistical projections as a control, and then see if layering observations about team chemistry would improve the precision of the predictions. So for the NL Central the PECOTA projections looked like this: NL Central W L
Cardinals 86 76
Cubs 85 77
Brewers 84 78
Astros 81 81
Pirates 79 83
Reds 78 84 My hope was that somebody would be able to say that, independent of talent, my ranking of the teams chemistry would be (made up ranking) Brewers Cardinals Cubs Reds Pirates Astros Then at the end of the year we could look at the actual results and see if the chemistry observation, layered onto the PECOTA projection, improved the precision of the predicted results. Obviously if the pre-season chemistry rankings match the PECOTA rankings there's no additional information and no test. The only response I've been given, from ABuck, I think is somewhat sarcastic in that he projected the teams with the best chemistry are the ones projected to be the worst in each division. However if, at the end of the season, none of those teams actually came in last place I'd be tempted to say ABuck is on to something. Like I think I've said before, I understand the arguments on both sides. I'm just trying to shine a little empirical light on the subject Yes, but how do you know what teams have good chemistry and which don't? It's a really subjective thing. Additionally, we can't observe each team day to day. Even if we asked the players, who's to say theyd give an accurate depiction of the team's chemistry?