Yes, it’s true that Point Shares tends to be off from ADP. It’s why I note that it does not make for a great draft sheet – particularly since hitters are generally valued greater than pitchers in drafts. I accounted for that in the 180/80 column which weighs up hitters and weighs down pitchers by adjusting the hitter/pitcher split from 152/108 (which would be an equal distribution of $260 amongst 13 hitters and 9 pitchers) to 180/80 (multiply hitters by 1.17 and pitchers by .74). That moves Votto to #5, Miggy up to #9, A-Gonz to #14, Fielder to #15, M-Teix to #20, and Howard to #29.

Now the differences between hitters get really small (e.g., Miggy is projected at .08 points less than A-Rod and .35 points behind Holliday) which is well below the likely margin of error. so I wouldn’t take this ordering as gospel. But I do think it’s very helpful to have an impartial guide to avoid personally overvaluing a player. Right now, I agree with Grey in preferring Miggy over Votto (I like Pujols better than Miggy though) and would consider A-Gonz and Fielder in late first round/early 2nd round. And I would consider M-Teix in middle 2nd round, Howard in late 2nd round.

But the impartiality of Point Shares helps me to make these choices. The reason I wouldn’t consider Howard any sooner than late 2nd round is 1) there is no sign amongst projections that he is superior to the others in R/HR/RBI and 2) there are clear signs that he is inferior in AVG. I have to balance this information against my perceived ‘safety’ in these picks – aka my gut (e.g., is Votto the real deal?). Grey obviously feels safer with Howard vs. M-Teix, A-Gonz, and Fielder. I see Howard as the AVG drain of the four, getting dangerously old for his body type (Fielder is 5 years younger). We’ll debate this before any drafts we co-run and we’ll figure out a blend of our two viewpoints.

FYI, part of the reason why 1Bs are higher in Point Shares than you might think (b/c of the averaging) is that it is set at 75% vs hitters at your position and 25% vs. all hitters. If it was set at 100% vs. hitters at your position, it would depress 1Bs further at the expense of shallower positions. I think there has to be a balance b/w the two as, at the end, you win based on team totals not positional percentage advantages.

Re: power/speed scarcity, I think it’s overrated outside of deep leagues (16+ team MLB, NL/AL-only). It’s true there are only so many 30+ HR and 30+ SB players. But you can build strength in power/speed via depth. I prefer to have power throughout my team with, ideally, at least one top SB guy. I find players with power tend to better on average in R/RBI whereas drafting a team of 20/20 guys can often come at the expense of R/RBI/AVG. I mean, if I can finish near the top of HRs in an 12 team NL-only snake draft where I picked Lincecum and Reyes as my top 2 picks, it’s easy in a 12-team mixed league. The depth in 1Bs (after the guys above, there are still Dunn, Morneau, Konerko, LaRoche…) would make me willing to invest a #4 pick in Tulo, hoping I can get M-Teix with the 2nd pick and, if not, I’ll get one of those other guys. My 1B/SS power will beat every other teams 1B/SS power…

The more I think about it, the more I’m curious about the methodology of comparing players against the average at their position. Looking again at your preliminary point shares, there are a lot of first basemen ranked well below where their ADP will be: Cabrera (13), Gonzalez (19), Fielder (21), Teixeira (28) and so on.

The objective of the game here is not necessarily to outperform the other teams position by position, but to outperform overall. I think Grey, without doing the math, has stumbled on to this concept by saying yes, 1st base is deep, but that doesn’t mean you don’t have to have one of the guys in the top tier. Of the 14 players forcast by PECOTA to hit 30+ hrs this year 9 are 1st basemen (and one is Chris Carter…no effin’ chance). Obviously the top tier is bringing up the average and bringing the points shares way down, but is it in the right proportion?

I think there’s a complexity to constructing the roster across 23 players that goes beyond how you’ve modeled scarcity. Something along the lines of, if you don’t draft the guys who hit home runs regardless of position there’s just not enough total power in the whole league at the other positions to compensate regardless of how much value you find relative to the point shares model.

Thanks again…just stumbled on this site a couple weeks ago. One of the best on the interwebs IMHO.

FanGraphs: 93/39/114/3/.271 – Nothing wrong here except that FanGraph fan projections tend to be higher. 8 1Bs have higher RBI projections are for 1Bs and none of them are Howard (Cabrera 139, Pujols 138, Fielder 135, A-Gonz 127, Votto 126, Dunn 121, Teixeira 118, Morneau 116). Everyone else has similar HRs to Howard except Morneau, Votto and Teix who are in the 29-35 range. And those are Howard’s strengths! His average is lower than all the aforementioned except Dunn. So, FG is projecting him as around the 8th most valuable 1B…

Marcel: 80/31/102/4/.263 – Marcel is historically low in their estimates so these are relatively high. Howard’s 103 RBIs are actually tops among all players and the HRs are tied for 2nd behind Pujols. So Marcel likes Howard.

CAIRO: 96/39/126/3/.267 – CAIRO also was kind to Howard as he tops their RBIs and is tied for 2nd in HRs.

So, aside from potential undervaluing by Fans @ FanGraphs, it’s not like Ryan Howard’s stats are that out of whack. Hell, I think anyone who drafts Howard would be more than happy with an average above .265 as well.

I think it’s how you value all these stats together. Point Shares is looking more scientifically than Grey does and it isn’t factoring in ‘reliability’ which Howard would score highly on. But looking at 2010 (http://razzball.com/2010-player-rater-for-12-team-mlb-leagues/), I see Howard was #58 among all players, #36 among hitters and 7th among 1Bs in 2010 Point Share value. The 6 ahead of him were Pujols, Votto, Miggy, Konerko, Teixeira, and A-Gonz. Dunn (if credited as a 1B vs. OF) and Fielder were just behind Howard.

Why? Well, Howard definitely is an average drain vs. the other players. M-Teix had a bad average year (-0.9 PS) which trumped Howard (-0.1) but the other players were between 0.7 and 1.7. Basically, having Howard vs. Votto or Miggy cost a team almost 2 points in AVG.

The next area where Howard trailed was runs. He was +1.0 PS which beat poor A-Gonz with his shitty Padre lineup and Konerko but was bested by Votto (+1.3), Miggy (1.6), M-Teix (1.7), and Pujols (1.8).

Now Howard is going into a year when he’s 32 (and he’s going to age bad – cough – Mo Vaughn – cough) and lost the best bat behind him in Jayson Werth. Pujols/Votto/Miggy are clearly better in my eyes. A-Gonz goes to a better stadium and lineup. M-Teix is due for an AVG correction. Fielder is due for some RBI luck. Morneau is due for a non-concussion year. Dunn is Dunn….

So, net-net, I generally agree with the Point Shares more than Grey. Howard is a safe pick but he really has no chance to deliver 1st round #s. I’d say he’s a safe late 2nd round pick that pairs well with H-Ram or Tulo. But my guess is that teams that draft Howard in the first 16 picks will generally underperform this year. But we’ll have to wait until end of 2011 to validate this with Razzball Commenter League data (2010 teams that drafted Howard performed slightly below average – an average of 7th place in RCL 12 team leagues vs. the ‘average’ of 6.5th)

@Francis: I’ll have 14 team up by end of the month.

@Steve: I’ll aim for a similar post around March 15th. Need a good amount of drafts before there’s reliable Average Draft Position data.

@Lava: Yup, you got it. You want to get as much value as possible out of every pick and, in general, teams draft heavier on offense than defense in early rounds. So if you want to create a draft sheet, I’d add 24 (or 20 for 10 team leagues) to every pitcher. Even closers. But you’ve also got to adjust to the draft room. If pitchers are being undervalued and there are a number of similar pitchers around but only one good 2B, you maybe reach a little for the 2B vs. draft a pitcher that might be around next pick.

You mention that the point share tables shouldn’t be used as a draft guide, and that you should aim for getting better than fair value for pitchers (1.5-2 rounds after where they would normally go).

I have a decent handle on auction values, but translating that to value in a snake draft always seems more fuzzy.

To synthesize, and most likely to betray my lack of understanding, the players should be drafted in the order presented by their rank, except for pitchers, which should be drafted at good (1.5) to great (2) value due to abundance and relative volatility.

Is this a fair assessment?

http://razzball.com/best-and-worst-values-by-position/

Are you planning to do it again this year?

@Eddy: Should have them all up by March 1st at the latest.

@Swagger Jackers: Nah, they are still helping. Could only hurt in AVG. But b/c I based everything on the average player, there are going to be at least 50% in the negative.

@Bubba: Fixed on Ks.

@Steve: Cool. Glad they helped!

I noticed that Ks are missing from the 12-team chart, just a heads up. (They’re included in the 10-team version.)

Though last year I didn’t use Point Shares, my venture into a 14-team and 16-team league this season will most definitely make them resourceful.

And, to get this question out of the way: when do you think you’ll have all the subsequent point shares up?