well I pulled the trigger on Milledge for Martin, I'm happy about that. the Milledge owner packaged him with $9 Volquez and $10 Werth for $12 Lincecum, which is obviously a killing for him.

Milledge has a pretty low trade value in my league right now...I think the point about unrealized potential being cheap is right on. I have an offer on the table to get $6 Milledge back for $16 Lidge and and $38 Sizemore, which is 2 guys I'm not planning on keeping at all (we have an 8 keeper max). I am super tempted to take it...it's good value wise but I'm afraid I could get more from other owners for those guys.

edit, and as we speak I got Milledge back for $16 Lidge and $9 David Price. kind of same deal as above....I think Price is fine but at $9 he is really no deal at all. and Lidge at $16 is no real value at all either. free Milledge imo

What does LPP say those guys were worth last year and Marcel say they're worth this year? I'd say paying a 30% premium on Sizemore would be better than having Milledge at a $2 discount. At the end of the draft, having a $30 Sizemore for $38 is much better than having an $8 Milledge for $6 with $32 leftover. My gut is telling me you got the third best keeper in the deal. It is/should be very difficult to get a player the caliber of Sizemore in a 16 team league.

I've got Milledge coming back at $6 and I probably won't keep him. The first factor is, pretty much no matter what, I'll be able to get him for $1 or $2 in my draft. Even if I thought he'd go for what he's worth, he likely won't be worth much more than $6. Even if I got him for $7, it would be worth it to reset his contract. I'd be in no-man's land at $8 or $9, but only booked at $10 lsat year and only gave $1.

It's too bad you don't have a $12 Lincecum

Lidge, less than $16, Price, nothing obviously, Milledge not much either. in reality may not be keeping ANY of them. I can keep 8 contracts and have

Kelly Johnson already signed at $7 and $10 in 2010
Cory Hart already signed at $7 and $10 in 2010
Hunter Pence already signed at $9 and $12 in 2010
Russell Martin already signed at $11
Matt Capps already signed at $12
Phil Hughes (ugh) already signed to $9 and $12, though this I may cut, taking a $6 cap hit in this years draft
I also have Cliff Lee who I can sign for $10 this year which is a pretty easy decision imo

assuming I don't cut Hughes, I then have 1 spot open for the following guys. if I cut Kelly Johnson for another $6 cap hit, I have 2 spots open to choose among

$51 Arod
$38 Sizemore
$6 Milledge
$22 Kemp
$10 Chris Carpenter if a miracle happens in spring training and he looks good

whats difficult to project is how many superstars will be available at the draft. b/c my league essentially started new last year, almost every top tier guy was drafted last year and can be kept for just $3 more this year. gut is telling me many will be. Pujols at $40, Hanley at $45, Reyes at $51, Wright at $51, etc will all be kept I'm pretty sure.

I hate giving up Sizemore for $38.

Hughes is a $3 decision, so that's a probable cut. Johnson is a $1 decision, so you've got to keep him. He's alright at $7 anyways.

What is your cap? If it's $260, then you keep A-Rod at $51 and Sizemore at $38. That gives you $151 spent counting the $6 for cutting Hughes.

Would the guy take Cliff Lee for Hanley? If that's an obvious yes, maybe you can force Hughes on him. That'd be a $16 Cliff Lee for a $45 Hanley. It's worth a shot. If Hanley fails, I'd probably try it with Reyes too. Lee is probably overvalued right now. He's a keeper at $10, but I'd rather have a superstar hitter at a position of need at book value.

Also, how is Capps a definite keep over Lidge at $4 difference? Lidge's K-rate is like double. That's huge even if it's only over 60 innings. That and the trust factor of remaining a closer.

my cap is $250...still I'm leaning towards both those guys

I'm actually trying to trade either of them + another marginal guy like Milledge for a Hanley or Pujols. in my league which counts offensive K's, Pujols is right at the Hanley/Reyes level

where are you guys valuing Lincecum at this year?

is he in the Johan/CC class?

edit...seems most people and projection systems seem to think so. wow he's really a $30 ish player? damn I hate myself

Sabthia, Lincecum, Peavy, and Santana, probably in that order, are the elite pitchers. You might throw Halladay in that group, possibly even ahead of Santana. So worst case, he's the #3 pitcher on the board, best case he's #1.

Thanks, again, for the help on this; it helped frame the problem and made the transition from a player ranking system to auction much, much easier for me. So, here's what I've developed so far, and I'm looking for any feedback, and definitely let me know, please, if you see anything fundamentally flawed with these calculations/assumptions. Thanks, and hope this helps some people.

Assumptions and calculations:
1.) Player value is most accurate when his statistics are figured as rates.
2.) Player value is most accurate when compared to the average of players at his position (this is a key difference from the replacement value model, as it only compares players active at a position, rather than comparing active players to their would-be replacement, or the worst starter) and of all players in the league.
3.) Player production is most accurately valued when considered as deviations from the mean production.
3.a) I also convert player production to a percent of the production required to expect a finish of one standard deviation above the mean (so, in a 12-team league, an expectation of about 10 points/cat), but this is not really necessary, it's just a way to see the approximate value of a player w/o consideration for position or category scarcity.
4.) Value of each standard deviation from the mean is determined linearly.
4.a) In my league with a 24-player roster, $240 cap and 12 categories, each category has a $20 value. Because the average player has a $10 value, being 0 standard deviations from the mean in any category should hold a value of $1.67 ($10 divided by the number of categories applicable to that player, so 6, in this case).
4.b) This is the biggest leap in logic, probably, but: I also assume that having a production of 4 standard deviations from the mean will *guarantee* the team a finish of at least one standard deviation above the mean in that category, so I value such production at the full $20.
4.c) These two points yield: value=4.75*standard deviations from the mean+1.67, for this league.
5.) From here, it's simple. Figure the value when compared to players at the same position, then value when compared to all the league's players, and average the two.

The result is a player value w/o inflation. Again, thanks for any feedback.

I don't have any formal education, so a lot of what you said is a foriegn language to me. It's been a lot of trial and error for me to get to this point. Perhaps I can improve my formula?


I believe this to be true, but this will be the first time I implement the strategy. At the very least, it should help us easily identify where the upside lies. Playing time still needs to be accounted for.

I sitll use a replacement model. I figure out the points gained based on the whole league, then set replacement level at the position. On my spreadsheet, I've got two values, total points and point value. The last starter at each position will have a point value of 0.00 even though total points may be higher or lower.

This would be the equivalent to gaining a point in the standings?

I believe so. They travel in packs, but those packs could be anywhere. Sometimes you'll be fighting for your life for a point or two in the middle, sometimes at the top.

I don't follow what is going on here.

This is another step that I'm pretty sure I don't do, or do differently, or maybe exactly the same.

I've been posting over at Last Player Picked. The price guide is pretty good for unadjusted values. As I'm sure you can all tell by the tone of my posts, I am a big fan of the stars and scrubs philosophy. As such, I believe there is a superstar premium on the elite players, especialy hitters, even in a non keeper league. This only gets exacerbated in keeper leagues where most or all of that money saved should go toward elite talent.

He was examining the 70/30 hitting/pitching split. Of course how much you spend on what is going to vary from league to league. This is what I wrote about my league;

This ties in nicely to the other post on inflation. The Price Guide reccomends that I have a 61/49 split, but I go 70/30. I’ll explain.

I’ll liken it to when I go to the grocery store to buy some Bran Flakes. I have to choose between the $4 name brand and $2 store brand. To me, they both taste the same, so the choice is pretty obvious. You would be amazed at how many poeple out there will say, “It’s only $2, what’s the difference?” Well, if I eat 50 boxes of ceral a year, that's 100 after tax dollars. Of course I don’t have an extra $100 in my pocket at the end of the year. I take that $100 and use it to buy premium items that I want, like steak instead of hamburger.

So it is in fantasy baseball. You don’t want to spend money on something you can get for free. In my league, to be a #3 starter you have to provide at least $8 worth of value. But I would never spend $8 on an $8 pitcher. Why? Because 11 of the top 37 starting pitchers in 2008 came off of the wiaver wire. This is not an unusual situation. So why pay $8 for a #3 starter when you can get one for free? On top of that, starting pitching values vary so wildly from week to week, it’s not all that uncommon to have a top 15 starter ($20) available on the wiaver wire!

So what I do is calculate the $values in a similar manner to the Price Guide, which comes out at exactly 61/39. Then, I take every starting pitcher worth $10 or less and move them down to $1. Then I take that money and spend it on elite offense. Thus, I can be “inefficient” and overpay for top flight hitting talent and still put together a very good pitching staff. The final tally usually ends up having an extra 9-10% shifted from pitching to hitting.



If I were to play in a redraft league, I might actualy end up spending $220 on offense and $40 on pitching...and expect to finish near the top in pitching!

Peru, does your ranking of SPs account for CC's twin 3500 pitch seasons or Santana's move into a presumably pitcher-dominant ballpark/improved bullpen?

All pitchers are risks, and the top 5 are no different. Here are my thoughts on the elite;

1. Sabathia-He's thrown a lot of pitches over the last couple of years, but he's a haas and at an age where he should be throwing a lot of pitches. I'm more worried about the move to New York. Betweeen the added pressure and less effective defense, at the very least I see a decrease in innings, and likely performance too. For me, he is the "safest" bet of the elite, which is why I have him #1.

2. Lincecum-He's the new Peavy and the best on an inning for inning basis. His issue is the rule of 30 and a huge increase in workload. Word around town is his mechanics are top notch, and I agree, so hopefully that mitigates some of the risk.

3. Peavy-Drop in Ks and an injury in 2008 are his risks. His got the worst mechanics of anybody at the top, so that adds to the risk. He's equal to Lincecum when healthy and had a light workload last year.

4. Santana-If everybody else has a red flag, Santana has a red flag and a siren. The decline was masked by some serious luck and his move to the NL. His walk rate has increased three years in a row and his K rate plummeted last year despite facing weaker competition. His fastball velocity is 2 MPH slower than 05-06. This reminds me of Maddux in 1996-1997. He had a "down" year, which was still probably top 5, then continued to be an elite starter for the next 5-6 years. He just wasn't the best as he got passed by Pedro, Clemens, and the Unit. Still, his luck-adjusted numbers do not support the year he had last year, so I will be vary wary.

5. Halladay-He was #1 $value and luck adjusted numbers last year, but I don't see 20 wins and a hold coming his way again. I also have to question the Ks since that's 70 more then he's had in any season the last five years.

All those guys have risks, but I just don't know who you can justify higher.

Cliff Lee-Was #4 last year just under $40, but his luck adjusted numbers just don't support that. His ERA was over a run lower than expected, and the wins...well, let's just say I'll take under 22.

Cole Hamels-He's got the same total innings issue of Sabathia, the same innings increase issue of Lincecum, and the same luck adjusted/drop in Ks issue. Of all the top guys, he might best be avoided.

Brandon Webb-22 wins is not on the radar for 2009. Going from Orlando Hudson to Felipe Lopez won't help.

Ervin Santana-Luck adjusted numbers were probably better than his real numbers which lead him to $30 in production. His fastball velocity was up as well. I can see him being the most underrated of the #1 starters next year, but I still can't see paying more for Santana than the top 5.

Maybe I do have what it takes to write a book!

so you think ervin slots somewhere into the $20-$25 range?I'm just a bit afraid of 05/06/07 ervin. Would you rather hve $5 ervin (and the option of $5) raises or $12 lincecum?

I would still take Lincecum. Yes, Ervin's luck supported numbers were better. But that just means he had an awesome peripheral year last year. Like where are you guys slotting Ricky Nolasco this year? (I know his luck neutral numbers aren't as good but its the same idea).

Lincecum is arguably the #1 fantasy pitcher this year and while there is some injury risk everyone like Kyleb and their mother who's analyzed his mechanics think they're superb.

Plus the Giants pen is better this year and their division opponents look worse as a whole, so I think he's due for some more W.

For 2009 Lincecum. I think the difference in my league will be about $35-$22. Santana could be justifiable if the future value is enough. You've got to come up with some kind of net present value model. My gut tells me one year of Lincecum is probably worth two of Santana, although if you can tack on extra years if he improves again, then it might be worth it.


I'm not that far down the list, but I'd imagine Nolasco has got to be somewhere in the 17-25 range? That's a $12-$15 pitcher in my league. Same ballpark as Cliff Lee.

really, no formal education ?

I actually do some of this in my head, no real #'s behind it though. I would certainly rather have a $10 value today than $7 today and $7 in 12 months...where I draw the line I'm not sure. my league is so inefficient and filled with so many goofs that it's not THAT hard to find and acquire next years great keeper....which tells me the gap needs to be pretty gap before I'll sacrfice value today

My league's informal Winter Meetings are tonight. There are a few possible rule changes to be ironed out. It's basically going to be lots of drinking and making fun of each other's teams.

I have no intentions to lay the ground work for any big trades as I have tons of cap space to address my needs in the auction. Hopefully I can pull off another steal like getting a 2nd round minor league pick (Which will end up being somebody like Hellickson, Tim Alderson or Derek Holland) for Rich Hill who's singed for $6 per thru '10.

I actually think what we're dealing with is the superstar premium in this case. Lincecum is saving $23, compared to Santana $16 then $11. If Santana were a $35 pitcher saving that amount ($19 salary), you'd probably jump on it.

On that note, I'm probably going to make another change to my formula that would give me a more accurate idea of just what exactly the superstar premium should be. I'm just waiting for a PECOTA spreadsheet to come out.

That got me to thinking about whether I should wait for the PFM to allocate PT or if I should just do it myself. Then I made a very disturbing discovery. The PFM does not adjust R/RBI based on team offense or batting order. That explains why the PFM overrates young players and significantly underrates established stars. The best example from last year is Bob Abreu vs Lastings Milledge. PFM had Milledge at $13 and Abreu at $9 with Milledge only having four less runs despite Abreu having a 20 point edge in OBP batting higher in a better offense. It also had Mileldge with more RBI, although it did predict a 40 lead for him in SLG with 4 extra HR. Still, Abreu had knocked in 100 five years in a row and batted in the middle of a good offense and while Milledge projected to bat #2, he is a lot less likely to stay in that spot if he struggles. Since R/RBI are 33% of the scoring in my league and 40% in most leagues, this makes the PFM and PECOTA almost worthless, IMO.

peru, thought you'd be interested to know rotoauthority is looking for part-time writers

Thanks. I'll have to look into this.

I've calmed down now over this PFM debacle. It's not as bad as I first though. Abreu seems to be the worst. The biggest problem seems to be with power guys who could be leadoff hitters. Kenny Lofton was Abreu's #2 comparable. It's probably more of a playing time issue than anything. That's one thing I've noticed. If you can project playing time, you have got a huge edge on your competition.

iirc playing time was a huge factor with PCM and milledge last year. Though some of the np fantasy writers loved milledge as well last year. Still yeah I suspect Pfm does overrate some young guys

Re: lincecum...am I paying too much if I give up $9 pence ($12 next year), milledge, and $38 sizemore? By value this is great for me but the thing confusing me is still the superstar premium for sizemore. Instead of keeping sizemore I would be keeping kemp if I did this deal

Do you have to keep 8?

I think I'd rather have Kemp ($20 in my format), Lincecum ($35), and $4 than Sizemore ($43). At those prices, Pence and Milledge are probably fungible. You ought to thow in Ian Kennedy as well!

well I ended up doing Pence, Milledge, and Kemp for Lincecum.

I don't have to keep 8; that is the max. anyways my rationale was:

I had about 10 guys who were keeper worthy. anything past 8 though is valueless, or only has trade value to other people. there was a chance I wouldn't be able to realize any value at all from players 9,10,11,etc on my roster. so I was really trying to package a few of them into 1 guy more valuable than any of them. and in my league, there aren't many of those "super keepers" out there. most are very hard or impossible to acquire - they already have been acquired by a few shrewd teams, so there's not that much left to buy on the market.

Pence is certainly a great keeper. worth about $13 or w/e in captured value to me this year. even if you value Milledge and Kemp each at a few bucks, they are nearly worthless to my roster at this point. and in return I got a Lincecum that captures me $20 or something like that in value.

I still have 6 guys under contract + Sizemore, Arod, and Cliff Lee. I am still going to try to do a 2 for 1, otherwise I will probably cut Phil Hughes. but my "top 8" certainly have more value after this Lincecum deal.