2. IMO you are misusing team pass attempts in these discussions. Bottom line, Irvin had 193 targets in 1995, compared to 199 for Bruce. That disparity is so small as to be irrelevant. IMO individual targets are clearly more relevant than team passing attempts in this discussion.
If you'll permit a little bit of argumentum ad absurdum...
Imagine two teams. One is the most pass-heavy team in NFL history, finishing the year with 400 pass attempts vs. 100 rush attempts. The second is the most run-heavy team in NFL history, finishing the year with 100 pass attempts vs. 400 rush attempts.
The pass-heavy team features three primary receivers. The first gets 167 targets, the second gets 133 targets, the third gets 100 targets. The run-heavy team has only one receiver on the entire roster, who accounts for all 100 of the team's targets.
Let's imagine that the #3 receiver on the pass-heavy team finishes with 80/1000/10 receiving, and the #1 receiver on the run-heavy team finishes with an identical 80/1000/10 stat line. Both finished with perfectly identical statistics across the board.
I would argue that the #1 receiver was *MILES* better than the #3 receiver, though. Literally the entire defense was focusing on him on every single passing play, since he was literally the only option. The #3 guy, on the other hand, got to feast on #3 coverages as defenses paid attention first and foremost to his two more-prolific peers.
This is an extreme example, (that's kind of how reductio ad absurdum works), but the exact same forces that are in play at ridiculous extremes like these are also at play to a lesser extent across the entire spectrum of possible situations. Michael Irvin carried the Dallas Cowboys' passing game to a much greater extent than Isaac Bruce carried the St. Louis Rams' passing game, and that difference, in my mind, easily makes up for 150 extra yards for Isaac Bruce.
(This is especially true if, like me, you view the "route run" as the best measure of WR opportunity, and not the "target". Counting sacks, there were 675 times in 1995 that a St. Louis quarterback dropped back, surveyed the field, and attempted to advance the ball through the air. In those 675 plays, Isaac Bruce managed to get open enough and catch the ball enough to accumulate 1781 yards. There were 512 times in 1995 that a Dallas quarterback dropped back, surveyed the field, and attempted to advance the ball through the air. In those 512 plays, Michael Irvin managed to get open enough and catch the ball enough to accumulate 1603 yards. Had Dallas thrown the ball 163 more times, I'd venture Michael Irvin easily would have gotten another 178 yards.)
1. I feel like we are having two different conversations here. Who cares if Bruce's season was an outlier? This tangent began by comparing Bruce's 1995 season to Smith's 2005 season... the top peak season for each. What Bruce did in other seasons is irrelevant to the discussion to which you responded.
2. Sure, it was a great year for several WRs. So? Fluctuations happen, whether due to talent (e.g., at CB1), rule emphasis, or simply a convergence of great players playing in their primes with other great players. 1995 happened, and it counts.
1. The reason it matters whether 1995 was an outlier gets to establishing what 1995 tells us about Isaac Bruce as a player. If one season stands out as wildly out of line with every other year of a player's career, it's a pretty good bet that that season is not very representative of his true level of play. See also: David Boston, Germane Crowell, Patrick Jeffers, etc.
(It also raises the specter of the possibility that the season was a result of random variance above and beyond the usual amount. Victor Cruz in 2011 hit on an absurd number of deep receptions. That year really happened, and we can't take those off the board, but his "true level of play" in 2011-- and overall-- is not very well represented by his 1500 yards that year.)
2. I'm not trying to say that 1995 doesn't or shouldn't count. I'm merely saying we have to adjust for the environment. You're familiar with era-adjustment, and I don't seem to recall you having any objection to it in general. When discussing Lance Alworth, it's worth making a note of the pass-happy AFL. When discussing great WRs of the '70s, it's worth bringing up the impact of the dead ball era. And when comparing Isaac Bruce's 1995 to Steve Smith's 2005, it's important to keep in mind that 1995 was, by a *mind-blowing* margin, the single greatest season for WR production we had ever seen prior to perhaps this year.
1. Agree submitting ballots before the regular season is over is a flaw in the process, which I have always felt, long before this conversation.
2. Through week 15 in 1995, here is where they stood: Carter 115/1292/16, Bruce 104/1571/12. I don't see that as a wash. Bruce had almost 300 more yards, despite the QB disparity already referenced.
3. I do fault the voters if their thought process was what you suggest here, i.e., well, it's close, and Carter was great last year, so let's give it to him again. I don't doubt that plenty of voters voted with that perspective, but it is wrong IMO. If anything, IMO the tie should go to the new young player over the established veteran great performer.
See, I think that's just being a smart Bayesian.
Let's think about coach of the year. If New England and Jacksonville both go 14-2 next year, who is likely to be the better coach, Bill Belichick or Gus Bradley? How about QBs. If Aaron Rodgers and Andy Dalton finish a season with identical stats, who do you think was probably the "best" quarterback that year, from an objective standpoint? If Antonio Brown and Marqise Lee both finish 2016 with the same receiving yardage total, and you had to take one of them to win a game with your life at stake, which would you be taking?
Bayesian inference is all about updating our priors based on new information. If you're asking me who the best ________ in the NFL is right now, I'm absolutely taking past performance into account. Aaron Rodgers is always one of the best quarterbacks in the NFL, regardless of what his numbers are right now. Bill Belichick is probably always the best coach in the NFL, regardless of what New England's record is, or of how much some other guy has outperformed expectations.
Similarly, Jerry Rice was probably always the best WR in the NFL through his entire prime, regardless of who led the league in receiving. (With a possible nod to Sterling Sharpe.) And Jim Brown wasn't a worse RB in 1962 than Jim Taylor just because Jim Taylor happened to rush for more yards that year.
Now, of course, it's tricky. I'm not saying Jim Brown should have been the 1st team AP All Pro in 1962, even though he was probably still the best RB. Taylor had nearly *50%* more rushing yards. That matters. Postseason honors should go to who performed better that year, and in 1962 it wasn't even close, Jim Taylor clearly performed better, (even if he probably wasn't the best RB in the world, overall).
But when things *are* close, I can easily see letting things like prior track record sway your opinion; moreover, I think that's a perfectly reasonable and acceptable process. And I can also see someone thinking after 15 games that Carter's 9 extra receptions and 4 extra TDs on a better team were pretty close to Bruce's 279 yards on a bad team that threw all the time.
Isaac Bruce had 39/550/3 while his team was down by 2+ scores in 1995. Cris Carter had 19/189/3. I doubt any voters thought about it specifically, but I'm sure there was a general impression that Isaac Bruce was racking up more statistics that just didn't matter, feasting on garbage time. Whether that's fair or not, we see it a lot, with players who have huge stats on terrible teams seeing their accomplishments minimized.
I've already said that I think Bruce was better than Carter in 1995. But I can see voters saying "well, Bruce had better numbers than Carter, but I still think that Carter was the better WR this year, anyway", especially after 15 games. Carter was the reigning 1st-team AP All Pro and the guy who had just set the single-season receptions record. Anecdotally, I think voters were placing a much heavier emphasis on receptions in the early '90s than they are today, simply because they'd never seen numbers like that before in the statistic.
Numbers are not a perfect reflection of performance, and they're an even more imperfect reflection of talent. If the NFL really wanted to just give all of its postseason honors to the guy who finished with the best statistics, they should rename it the Stat Line of the Year award. The statistics should be of primary performance, (especially when adjusted for context), but statistics only matter as a proxy for general performance. And like I said, when things are close, I think it's pretty reasonable to err on the side of the known quantity.