Jockless Joe had 4 picks highlighted as best in their round, while the Naptown Bombers had three. I, the Greenwood Cubs, only had one (Hidalgo), While Serpico, JoshBooty, and Plate O'Shrimp had none.

I wouldn't want to question your objectivity, but you overlooked my absolutly brilliant picks in the 3rd (Rivera), 8th (G. Anderson), and 18th (D. Hermanson).

This, however, will not earn you the hate of the rest of the league. If you had posted the worst picks in each round, that would have earned you the hate of the league.

Aw, c'mon, since you haven't spun everyone up at this site with a Clay Aiken reference, why not piss people off with listing the worst fantasy picks? You will piss everyone off eventually, why put it off?

I am also all for a 'worst pick in each round' addition to this article. I'm in the 2nd league, but I'm hoping for one of these types of articles in that league too. Should get some good banter going anyways. Did you find anyone to switch with Will to get him into the 2nd league? When do we find out draft order for that second league? Do you draft 19 starters and 3 bench guys? If you draft a guy on the DL, then right after the draft can you pick a guy up in place of the hurt guy and then slide the hurt guy over to the DL? Brimming with questions.

I'm in the league as the Dusty Busters. I would like to see that "Worst Pick of Each Round" breakdown as well.

Ok, first off, I've been exposed on my claiming to have the best picks. I don't think I had the best overall draft, though.
Rick, I would say that Garrett Anderson was a tossup with Berkman and Rivera was right behind Sheffield on my list.

In regards to worst pick, please feel free to do your own and list them, as I just don't have the time, currently.

On the subject of the DL, players who will not start the season have not been classified that way, yet. I took a couple players hoping to pick up another, but I haven't been able to yet. I think the draft order isn't chosen by Yahoo until right before the draft. And yes, Will was able to get into the second draft.

I'm playing as the Chicago Fruit Bats. Nice draft, everyone. Put me down in favor of a worst pick per round post, too.

Scott: I think that's a pretty good list, though I noticed it was a bit your-team-heavy as well. That's no problem... you had a good draft. I'd put Kevin Millar above Cliff Lee in the 17th round, and Placido Polanco neck and neck with Todd Walker in a 22nd round that was all about value 2nd basemen. Maybe Wily Mo over Wells in the 18th, too.

Hey .... I only suggested that if Will or yourself wanted to be hated by the entire league, that a worst picks list would do it. I'm not going to pick on anybody else. I know the ones I thought I screwed up on ... Matt Morris ahead of Kerry Wood for instance.

And Scott ... I thought about Berkman, but figured he's hurt right now, so I moved him down the list.

Also ... in Eric Altermans column on MSNBC there is a listing from some guy from UNC who says the ten greatest sluggers are

Ruth
Ted W.
Rogers Hornsby
Barry Bonds
Lou Gehrig
The Mick
Stan M
Ty Cobb
Jimmy Fox
Say Hey…

The guy who said that 4 guys were ahead of Bonds was wrong about his count, but was correct in his statement that they are all dead.

That UNC guy would be Michael Schell, I think - a notable biostatistician (well, more than me, anyway), and author of at least two books on using statistical models to rank hitters/sluggers across eras. At least in the "Hitters" book, his models seemed reasonably solid to me, at least fairly difficult to poke big holes in, but it's also hard to see what the point is to identifying the best hitter for average, ever, as opposed to the best hitter period. I expect the "Sluggers" book to be more of the same - anyone seen it yet?

I'm in the league, as well, playing as the Penzance Pirates... Good luck to you all.

OK, I'll bite on the "worst picks" thing, sort of. In an attempt to distance myself from the results, I used Prospectus' Player Forecast Manager (basically PECOTA) to rate the draft for value, as if it had been an auction. Details at the end of the post.

For each round, I let the infernal machine identify the player with the most value and the least value. Perhaps I should something gentler, like "biggest stretch."

I disagree frequently, of course, but it was fun to play with, and I do think teams incorporated the impact of the SABR-categories on the values they assigned to players differently. For instance, I wouldn't have picked some of the low OBP speed demons if they'd been there in round 20, but Crawford, Pierre, and Posednik (for instance) were long gone by then.

13 of the 16 teams got the "most value" in at least one of the 24 rounds (Chicago Fruit Bats led with 4 such picks), while 11 got the "biggest stretch" in at least one round (2 teams tied with 5 each). I made no attempt to deal with draft slot, so it's certainly not fair for that reason among many others. What I'd be curious to find out is whether these results mirror the degree of reliance on PECOTA or the PFM to prepare for the draft.

What follows is Round #, then most valuable and then biggest stretch in the round, with pick # in parentheses. So, for instance, in round 1, Bonds was the player selected with the most value according to the PFM and was picked fifth, while Beltre was the player with the least expected value and picked 13th.

Round:Best (pick #), Worst (pick #)
1:Barry Bonds (5), Adrian Beltre (13)
2:Adam Dunn (21), Carl Crawford (29)
3:Keith Foulke (43), Juan Pierre (47)
4:Armando Benitez (54), Carl Pavano (59)
5:Troy Glaus (75), Rich Harden (67)
6:Brian Giles (83), Jose Reyes (89)
7:Justin Morneau (105), Chone Figgins (104)
8:Lance Berkman (119), Jeremy Bonderman (128)
9:B. J. Ryan (133), Luis Castillo (134)
10:Eddie Guardado (157), Jody Gerut (145)
11:Preston Wilson (165), Angel Berroa (173)
12:Jeff Bagwell (182), David Wells (191)
13:Dustin Mohr (206), Dave Roberts (208)
14:Frank Thomas (216), Brian Schneider (222)
15:Erubiel Durazo (227), Bill Mueller (235)
16:Mike Lieberthal (248), Tony Womack (244)
17:Rafael Palmeiro (259), Ryan Drese (262)
18:Geoff Jenkins (287), Chris Burke (286)
19:Jose Cruz Jr (289), Ben Molina (304)
20:Reggie Sanders (306), Mark Ellis (318)
21:Garrett Atkins (322), Cesar Izturis (326)
22:Jacque Jones (339), David Bell (347)
23:Esteban Loaiza (359), Terrmel Sledge (364)
24:Tino Martinez (377), Alex Cintron (382)

Apologies if I've mangled a name here or there.

Finally, based on simply adding up the $ values assigned by the Player Forecast Manager for each team (at the end of the draft), three teams are well ahead of the pack (including mine, of course, since I fully admit I relied a lot on the PFM), and one is well behind. I'll be very interested to see if that holds up. Of course, trades, releases, and free agent pickups are already happening so ...

It was a fun way to spend a couple of hours, though. Play Ball!

Details: I used the Player Forecast Manager (March 9 version), set up with our eight categories (W, SV, ERA, WHIP, OBP, SLG, R, RBI), a 19 player and 16 team league setup (i.e. ignoring the 5 bench players), incorporating the positional adjustment, a "moderate" disposition towards valuing players, and using a $260 budget (minimum $1 bid) weighted $165-$95 in favor of hitters, which seemed to fit with the fact that we spent (I think) 101 of our first 160 picks (63%) on hitters.

'Twas I, Illini Nation, who took Will's spot by moving from the second league to the first one. Glad I did because this looks like a great group. Whoa, looks like I had three of the worst picks according DrTom's PFM analysis. That's ok. Last year I nearly got laughed out of the draft room for picking Aaron Rowand. Sometimes my gut is actually right.

I'm proud to say that (I think) the only one of my picks on the bad pick list is Terrmel, who I picked for chemistry because I love him.

Dr. Tom

I must defend myself, being the owner who picked Carl Pavano at #59. I ranked my players using winshares, which I got from the Hardball times. Carl Pavano was ranked 6th among pitchers, tied with Roger Clemens.

I guess that means I followed the PFM sheet pretty well, then. It was one of my main tools in the draft. I think PFM doesn't like Bill Mueller in round 15 because it's not aware that he's 2B-eligible in our league.

I went for a lot of high-end closers in this draft, figuring that with only a 7-inning minimum, a closers first strategy would give me a good shot at three of the four pitching categories every week. We'll see how that's going to work out.

I've been in fantasy leagues for 10 years now and let me already remark this is the best one I've ever been in. Dr. Tom, thanks for all your data and you took a very smart approach using statistical data to decide who made the worst picks.
Obviously, using my method of choosing the best player in each round was very flawed, as I wouldn't have picked the player in the first place, if I didn't think they were a good pick. I would be interested to see the best player in each round, using your data, Dr. Tom. Also, I like your name, as it reminds me of one of my favorite college basketball coaches of all-time, D.T. Davis.

Please forgive this ridiculously long post. I just got on a roll.

First, to clarify, I'm just a subscriber to BP - I certainly am responsible for nothing more than the idea of compiling the forecasts here. Pre-draft, I took the PFM data to make a first list (beats Yahoo's, anyway), then moved people up and down several times, some quite a bit.

And I certainly don't agree with all of the results I posted. For instance, I loved the Rich Harden selection at the time (listed above as a "worst" pick) and really wasn't that thrilled to wind up with Dustin Mohr (even though it's one of my "best" picks). Like Scott, I thought getting John Buck in round 23 might be the best value pick I made in the last 10 rounds, especially if a one-catcher team suffers a bad injury. We'll see.

A few specific reactions:

10 - Illini: I'm certainly not laughing about Beltre - although I'm guessing you'll agree that there's some risk there. I had him going in the 5th round on my list (I was risk averse early), but felt there was no way he'd make it past round 2, so late 1 isn't surprising. Berroa and Womack I have no use for, and neither does this pass of PECOTA in our categories:
Berroa: .310 OBP / .410 SLG / 70 R / 66 RBI
Womack: worse in all four categories

11 - Johnny (Dusty Busters) - For me, Terrmel Sledge ranks right up there with Whammy Douglas, Drungo Larue Hazewood, Salome Barrojas and Garth Iorg. Besides, no one who took Yhency Brazoban in round 24 can say anything about grabbing a name late.

12 - Rick (Greenwood Cubs) I love Hardball Times (more smart folks writing well), and enjoy the notion of Win Shares very much. Yet another idea where Bill James got me thinking deeply about something - lots of fun. And, of course, spirited defenses were exactly what I was hoping to spark with my previous post (and, I suppose, this one). But using 2004 Win Shares as a tool for picking players for a 2005 draft? First, WS incorporate so many things that we don't count in this league, like adjustments for park and league, and many other important things (strikeouts, unearned runs, defensive support, homers allowed). Also, WS clearly value Saves more rationally (when compared to real baseball) than we do in this league. That said, Saves are one of our 8 categories, and with small inning limits, relievers are necessarily at a premium, relative to their "real" value. Not to mention that WS also includes batting and fielding for pitchers, although it hardly matters in this case. But the real problem is that while I agree that Pavano had a stellar 2004, I don't think there's much chance that this year will be anywhere near as good. PECOTA agrees - the PFM as I set it up ranks Pavano around 95th among all pitchers (starters and relievers). That may be way too low, (and in fact, I had moved Pavano up a bit on my draft list) but the prediction seems about right at 12 Wins 190 IP with a 4.64 ERA and 1.38 WHIP. I think he's a league average pitcher going into a bad situation (will Bernie really play center?) But I've been a Mets fan for 30+ years, so you might want to take any Yankee statements I make extra salty.

13 - Eric (Fruit Bats) Under my little rules, the PFM likes Mueller less than I do, and does rank him solely at 3rd, behind people like Casey Blake, Aaron Boone, Chad Tracy and Fonzie. At 2nd, he looks like as good a bet as, say, Placido Polanco, to the PFM. I'd hardly defend that pick as "worst" in that round, but I'll spare you my personal views on that issue, as I'm afraid this "lecturing" post has already irritated some people in the league.

14 - Scott - The list I gave in post 9 does provide the best player selected in each round as well as the worst, i.e. Bonds, then Dunn, then Foulke, etc. And I'm such a non-college hoops fan that I first associate Tom Davis with Al Franken. But hey, I'll take good feelings wherever I can get them.

I guess I should mention that the PFM (new versions come out regularly) is available to BP Fantasy or Premium subscribers, and is cool. End of commercial for company I have no stake in. Cheers to you all!

Scott,

RE: Ben Sheets,

By 'inspired' I assume you mean 'brilliant' or perhaps 'canny' instead of 'stupid'. I might not have drafted Mr. Sheets if I had actually been present to draft rather than relying on the autodraft; I did take advantage of the pre-draft ranking option and mostly eliminated players as opposed to ranking all the good ones. Obviously though, I did rank Mr. Sheets as my overall #1: all I can is that decision was part of an overall beer-induced strategy that I am not at liberty to divulge completely at this time.

I will just point out that the number two pick Johan Santana and Ben Sheets are very similar (at least where 2004 stats are concerned:

MLB Starter ERA:
Santana 2.61 (3rd in MLB)
Sheets 2.70 (4th).

MLB K's:
Santana 265 (2nd)
Sheets 264 (3rd)

MLB Whip:
Santana .92 (2nd)
Sheets .98 (3rd)

Now granted Johan won a pile more games than did Mr. Sheets, but I don't think the Twins will be quite as good again, I also don't think he'll get to pitch 5 times vs. the Royals again and only once against the Tribe.

Santana's a fabulous pitcher, but I simply felt that Sheets was the best bet for a great year out of all the starters in MLB.

I guess we'll see. Overall I'm pretty pleased with my starting corps:
Sheets
OLPerez
Greinke
Smoltz
Bonderman
Carpenter
Bush
Rusch

Only Bonderman had an ERA over 4.00 or a whip higher than 1.23, of course Ryan Klesko might lead my team in homers.....

Dr. Tom

Not being a subscriber to BP, I'm at a disadvantage when it comes to PECOTA. I have to go with whatever stats I can get for free, and my gut instinct. Winshares is free, and my gut tells me that going to a team that won 101 games last year, even with Bernie Williams in center field isn't what I'd call a "bad" situation.

Well ... unless I was a Met fan.

And continuing about your original list, Chone Figgins, qualifies at three positions, making him more valuable than traditional stats would tell.

Scott in Illinois -

I liked the Sheets pick, it just surprised me that he was first ahead of BB, AP, and others.

Dr.Tom

Out of curiosity, who were the three teams that dominated in the total $ value for player forecasts, and who was the bottom of the barrel???

Re: Comment 18 - Sorry. I had the day off yesterday, so I could play around with that stuff, but I don't have access to it at work. I know mine was one of the top 3 teams (Penzance Pirates) by that measure and I think the team that picked Bonds was as well, and I also remember that the range of team values was somewhere around $100 (minimum) to $350 (maximum), but that's it.

Big Caveat - Please note that I attach absolutely no significance to these results.

I used the PFM heavily - as a result, my team looks good. Others who didn't, won't look as good.

Teams near the top clearly invested substantially in hitting, relative to some at the bottom. PECOTA is VERY conservative regarding pitching. I think that's a good thing, others don't. Teams who invested more in pitching won't look as good.

I am certain that auto-draft would make some teams look worse than they would if they had been able to draft live, and I am also certain that, while Nate Silver is very smart, the PFM is (a) dead wrong on a lot of people and (b) not taking into account lots of important information relevant to our game.

And, I have no idea how things will play out head-to-head, as these #s probably have more to do with depth and avoiding the big mistake than will our daily matchups.

Just because something can be quantified a little bit doesn't mean we know anything at all about the future...

In that spirit, here are the Pre-Season Draft Rankings (via PFM), using all 24 picks ($ value, first round draft position, and their three most valuable drafted players [with drafting round underlined, as if it was a link])

1. Penzance Pirates ($367, 14th, Thome 1, Edmonds 2, Hoffman 5)
2. Naptown Bombers ($364, 5th, Bonds 1, Gagne 2, Matsui 3)
3. Chicago Fruit Bats ($348, 11th, Helton 1, Abreu 2, Glaus 5)
4. Dusty Busters ($283, 12th, Dunn 2, Dotel 5, Prior 1)
5. Branyan Bashers ($275, 2nd, Hafner 2, Santana 1, KRod 3)
6. Jockless Joe Jackson ($260, 9th, Sheffield 3, Unit 1, Pedro 2)
7. von der Ahe Spiders ($247, 7th, Manny 1, Cabrera 2, Huff 3)
8. Bay Area Laboratories ($226, 10th, Teixiera 2, Berkman 8, Tejada 1)
9. Josh Booty Experience ($196, 1st, Lidge 2, Sheets 1, Andruw Jones 5)
10. Reunion Busters ($176, 8th, Vlad 1, Big Papi 2, Mora 3)
11. Plate O'Shrimp ($169, 15th, Soriano 2, Delgado 4, Beltran 1)
12-13. Serpico ($164, 4th, ARod 1, Guardado 10, Aramis Ramirez 3)
12-13. Illini Nation ($164, 13th, Beltre 1, Schmidt 2, Drew 3)
14. Student Loan Sharks ($162, 3rd, Pujols 1, Wagner 4, Schilling 3)
15. Greenwood Cubs ($157, 6th, Rolen 1, Rivera 3, Jeter 2)
16. Purpura Purveyors ($98, 16th, Chavez 1, Oswalt 2, Aquino 14)

Regarding the Purveyors, there are a few hitters on his draft list (especially he went on autodraft) that PECOTA hated, like Jody Gerut, Dave Roberts and Alex Sanchez, that clobber the overall score unreasonably. For various reasons, picking pitchers PECOTA hates doesn't do quite as much damage. Ignoring the bottom 5 players each of us drafted would put the Purveyors on a much more even footing, I'm sure.

Cheers!

Dr. Tom- Thanks again for your exhaustive breakdown of our draft. Really interesting to me. One question I would have is does PFM grade just players on just the stats we use, as I would guess they aren't particularly interested in runs, rbi's, saves, wins. I would have loved to create a league based on more important stats, but I had the categories what Yahoo presented me and wanted to have more than a 4 category league (OBP, SLG, ERA, WHIP). And how does PFM deal with park effects. Just interested.

Hey, Scott - Yes, the PFM lets you specify the categories you want it to consider in setting values, and I did use only our eight categories.

Re: park effects. Essentially, the PFM does take park effects into account, so that, for instance, if the Rockies traded Helton to the Marlins for Delgado tomorrow, Carlos' predicted PFM value would increase and Todd's would decrease in the next updated version of the PFM. Details follow as I understand them...

I believe that the PECOTA estimate that fuels PFM deals with park (and timeline) effects by first translating everyone's stats to the "same" park and time, then comparing them to other, similar players that have also been translated. Next, PECOTA produces a series of "translated" projections meant to describe the player's likely distribution of stats based on how comparable players performed in subsequent years. PECOTA then translates those values back into the player's likely park context for 2005 (obviously making some broader guesses as to park effects for some teams, like the Nats). So what fuels the PFM is essentially a weighted average of the distribution of predicted player outcomes in the park.

I could be wrong about some details here, but I'm a statistician (had you guessed?) and I've read what Nate Silver and the rest of the folks at BP have written about that stuff pretty carefully, so I think I have it. Obviously, it's a lot of work.

Thanks for reading, DrT

Any breakdown on the 2nd draft held Saturday night? Predictions? Best/Worst picks? I know I'll be listed in one round as I picked the wrong Luis Gonzalez!! Ooops.