Snodgrass'Muff said:
Short of a fire sale, I would be surprised if they finished below .440. Moving Koji would hurt, but I'm expecting better performance ROS from Panda, Hanley, Papi and Porcello and more growth from Bogaerts, Betts and Rodriguez. The mitigating factor is Buchholz seemingly being lost for the rest of the season, so I don't expect them to have a monster two month stretch, but they could go .500 the rest of the way.
And I didn't ascribe motivations to your thinking and I certainly didn't assume that all simple thinking is on the other side of the debate or that all people on the other side are doing it. I very specifically cited two posts as examples of what I was talking about and said a lot of people are doing similar things when posting. I think that bears out with a rereading of the most active threads on the main board right now over the last few days.
So you are predicting that on average, 39 out of 40 players have under performed each year when compared to projections? Because that's what you started with and that kind of hyperbole is what I'm arguing against. If you are amending that to suggest that the split would be more along the lines of 90/70 over 4 seasons I'd be interested to see the results of a look like this, but as it's your position, I'll leave it to you to do the legwork. You'll also need to come up with an error bar for it as it's not really fair to draw the line at the exact comp as being ~5% above or below is probably just normal variance and should count as a "heads.".
It may turn up less than 70 seasons of positive results, but I wouldn't be shocked if it turned up more. This year alone I'd bet on Bogaerts, Betts, Koji, Eduardo, Holt, Pedroia, Miley, De Aza, Hanigan, Buchholz, and Tazawa are all good bets to be in that margin for error or above it and this is probably the worst season of the four. You're likely to have a lot of players from the 2013 roster coming up heads as well, so that's a pretty good start. When you consider that the 2012 squad was 53-51 on July 31st of that year, there are very likely a good number of players who will also come up heads for that season as well.
And that doesn't mean the front office is above reproach or that they haven't made any mistakes worth criticizing, but the last 3.5 seasons haven't been as bad as some here are suggesting.
Okay, if you can point me to a way to get pre-season Zips projections for previous years, I'll do it. For 2015:
Six batters so far have performed more than 1.0 WAR under projected (projected was adjusted to 100 games, but the 1.0 WAR difference is over that 100 game time frame, not converted back to a 162 game total): Sandoval, Ramirez, Castillo, Napoli, Nava, and Craig. Four batters have performed between 0.5 and 1.0 WAR under projected (corrected to 100 games): Leon, Ortiz, JBJ, and Pedroia. Betts (0.8 WAR above projected), Bogaerts (1.1) and Holt (1.4) are the only hitters that exceeded expectations thus far.
I'm leaving out Victorino, Vazquez, and Hanigan for being injured - not because they are missing too many games to underperform, but because I'm willing to put injuries in the luck column. Castillo, JBJ and Craig are left in - their underperformance is due partly to their underperformance at the major league level and partly to the front office's unwillingness to play them.
Clay Buchholz (+2.4) is the only pitcher who's been more than 1.0 WAR away from projected in either direction. Porcello, Barnes, and Ogando haven been between -1.0 and -0.5. Masterson, Mujica, Breslow, Hembree, and Wright were between -0.5 and 0.0. Ross, Miley, Kelly, Varvaro, and Layne were between 0.0 and 0.5. Uehara and Tazawa have been between 0.5 and 1.0.
Please let's not have the WAR argument again. I'm assuming that over 29 different players the noise evens out (that WAR is an unbiased estimator). So we have in 2015:
6 players more than 1.0 WAR below expected
7 players between 0.5 and 1.0 WAR below expected
5 players between 0.0 and 0.5 WAR below expected
5 players between 0.0 and 0.5 WAR above expected
3 players between 0.5 and 1.0 WAR above expected
3 players more than 1.0 WAR above expected
By my 1.0 WAR criteria, that's 3 heads and 6 tails. If you project it out to a 162 game season, it's 6 heads and 14 tails. Totaled together, all of those players have been 12.6 WAR below projected so far. If I had some estimate of variance for Zips I'd be happy to run that comparison, although just a simple non-parametric rank-sign test has a p<0.0001 that the projected and observed performances aren't different - that is, random.