Actually, this reminds me a lot of some angry emails I got about Dan Murphy's projection.

The same arguments - OH MY GOD CANT YOU SEE HE CLICKED! And HES CLEARLY A SPECIAL PLAYER THAT NERDS DON'T UNDERSTAND (Not you, Sam). Turns out he wasn't a 313/397/473 guy after all!

You nerds don't understand the power of the human spirit.

Brock clearly has no less than a 1 in 100,000,000,000 chance of hitting 60 HRs next year.

Seriously, you guys should see some of the e-mails I get. If I wasn't already behind on things I need to do, I'd do a regular Reader Mailbag feature.

The less said about the whole Daniel Murphy thing, the better . . . .

You know, now he's going to hit .290/.350/.475....

I remember one of the old Bill James abstracts, where he mentioned that he projected someone to hit .220, the dude hit .280, James said he was wrong, but the next year (and every year after) the guy hit .220 or less...

Then came Rob Deer, who hit .232/.336/.494, and James said, "nope, not falling for it, if he could hit .232 he'd be useful, but he's going to hit .180...

A couple years later James said, OK I give up, he improved, he's better than he was in the minors, Deer immediately cratered and hovered around the mendoza line for the rest of his career.

Brock's 265/344/334 is actually a bit better than I expected. I don't know if that's a bad thing or not.

first of all fausto carmona is on great pitcher look at his wins and era and 2007 with cleveland. so in one year became on the a-list of pitchers and fausto is in demand now, now he is going to the hall of fame with home run king adrian beltre. and he also worked with sabathia and cliff lee who was great by the way. and tim belcher on his fastball. so he worked with alot of great pitchers. and was great in the minors. and fausto was the best thing on the team and the critics agree. so fausto is a way better pitcher then matt cain. im responding to the post from dan about at least matt cain has some experience under his belt but fausto is hotter and a better pitcher and he is getting alot more awards then matt. so please did you see matts last game against the padres. need i say more the game sucked as hell and even the critics said the worst pitching of the century. it only had 20 thousand fans in the stadium now thats sad.

The early part of this thread reminds me of one of my favorite mgl rants (of which there are many):

I’ve mentioned this before, but if a guy hits .850 (in a league where the average player hits .750) in his first year (or in any year, and that’s all we know), and he gets better in talent (say by 20 OPS points), how many teams do you think understand that he is supposed to hit around .820 the next year, a “regression” of 30 points?

For those of you who are just casual readers, let me repeat that: If a player hits .850 (OPS) his first year and we KNOW (G-d comes down and tells us) that his true talent is going to increase the next year by 20 points (maybe because he is on the up-slope of his aging curve, like Martin presumably is, but it doesn’t matter why), then we expect him to hit around:

.820 the next year. Yes, we expect that a player who gets better in true talent will hit 30 points less than he hit the year before!

You know, I never forgot the lesson.

dan i hope you charge tropic of cancer lots of money for the dmb build discs

I saw the Bond projection in 100 and thought "Barry would be better than that"

Bond's odds of hitting 60 home runs in 2010:

1 in 1,751,072,931,543,390

Dan, you need to watch him play!! The real odds are about 1 in 32.

I remember one of the old Bill James abstracts,

Or BPro who kept calling for Marcus Giles to get more playing time. He gets some and kinda stinks and there might have been rumors of attitude problems too. So for the 2003 BPro, they finally give in and admit Cox is smarter than they are. And of course he goes out and rips the cover off the ball.

Seriously, you guys should see some of the e-mails I get. If I wasn't already behind on things I need to do, I'd do a regular Reader Mailbag feature.

Getting angry at the maker of a projection system because you don't like a projection of a particular player seems awfully strange to me -- but it's been a strange thread.

Keep 'em coming Dan; can't wait to see Vernon Wells's projection! (groan)

Letting Garko go was sheer idiocy, given the Giants' weaknesses at 1B and the outfield corners. (On a related note, Garko should probably have some sort of DMB defensive ratings in LF and RF.)

Is 'sad face' a rating?

Better than the ROFL smiley for Adam Dunn, I suppose.

Dan, any idea what Tony Pena Jr is going to do? I'm one of the few people who saw his relief appearance a couple of years ago and I must say I'm intrigued. He could end up in the pen if a lot of things break right.

ZiPS reports a lot of wacky things. Bond's odds of hitting 60 home runs in 2010:

1 in 1,751,072,931,543,390


So, you're saying there's a chance!

ZiPS reports a lot of wacky things. Bond's odds of hitting 60 home runs in 2010:

1 in 1,751,072,931,543,390

Dan, how do you get that result? Clearly you're not running 1751 trillion simulations of the season. If you are running simulations, I can't imagine a PC could handle more than 1 million and still deliver results in a timely fashion.

Why are you getting such "precise" results?

Hey Dan-thanks again for ZiPS.
I've been mulling this over for a long time and I wanted to ask you-how much effort do you put into predicting W's? Does ZiPS project it based entirely on the projected peripherals and IP or is the team offense and defense included somehow? I've always operated on the assumption that you wouldn't devote too much energy to such an unpredictable stat, but I'd love to know how much does go into it.

Along the same lines, does anybody have any thoughts on what stats you would look at to predict a high (or higher) winning percentage? I'm trying to rank teams based on how much they're projected to help their SP's get wins but there are so many variables. (Obviously this is a fantasy question-I'm in a 5x5 where a lot of us tend to value players similarly so I'm looking for some new tricks with some of the luckier categories).
Thanks again.

According to San Fran CBS:

"SOURCE: GIANTS SIGN AUBREY HUFF
Quick blog here...got a tip from a source that the Giants have signed left-handed first baseman Aubrey Huff to a one-year, $3 million contract. That will almost surely set the Giants line-up for 2010 with Pablo remaining at third base. Kim Coyle reported yesterday that the Giants were still talking with Adam LaRoche, but my source now says that deal is dead, but can't say for sure. Huff stumbled last season to .241, 15, 85, but if he can return to his numbers of 2008 (32, 108, .304) this could be a steal. Stay tuned."

DeRosa to LF?

Why are you getting such "precise" results?

Well, it's not really that precise, but if there's a chance that, for example, so-and-so develops into a "true" 40 home-run hitter, there's a range of probabilities around that (which ZiPS looks for), even if a 60 home-run season is never the mean projection.

I've been mulling this over for a long time and I wanted to ask you-how much effort do you put into predicting W's? Does ZiPS project it based entirely on the projected peripherals and IP or is the team offense and defense included somehow? I've always operated on the assumption that you wouldn't devote too much energy to such an unpredictable stat, but I'd love to know how much does go into it.


Wins are very volatile.

Previous years run totals regressed towards mean is a pretty good predictor of team offense. Number of seasons tend to be predictive (bullpen quality is a big fact, obviously, but decisions are also a residue of endurance, which tends to be predictive).

Win totals for starters work similarly to teams, though the exponent is higher.

Saw Neal in the Fall League at Strasburg's first start and he had an outstanding game.

1. Looks like he's got a very strong throwing arm.

2. After Strasburg clowned several hitters backdooring that slurve thing of his, Neal kept his hands back and roped it into left field for a hit.

3. Made a couple of outstanding running catches in left and nearly made a third.

Now one game is one game and I'm not about to change policies on evaluating players from one game. But just from a standard stat geek perspective, when you see how much he hit, your first concern would be his defensive ability. And it sure looked like that it wasn't an issue for him.

Sabean really confuses me as his organization seems to do all the difficult things well, and can't seem to do the easy thing which is put together a reasonable offense. A little known fact was that he was scheduled to be one of the guys at the infamous stats/scouts roundtable but pulled out. Pat Gillick was there as an observer but didn't participate.

Sabean really confuses me as his organization seems to do all the difficult things well, and can't seem to do the easy thing which is put together a reasonable offense.

I think this can be explained in a very simple way. Sabean uses BA instead of OBP. Sometimes he will end up getting guys with a decent OBP because they have high batting averages, and sometimes he will get guys with decent walk rates by sheer chance (Durham, Burks, a few others over the years).

Notice also that the Giants pitching staffs over the last several years have been near the top in walks allowed - no matter if their pitching has been good or bad overall. I suspect, but cannot proove - that there is an organizational philosophy around "nibbling" rather than challenging hitters.

It's obvious on the offensive side, that the organization values aggressive hackers.

I have a feeling that having Bonds getting thousands of walks, often in weak lineups has colored his position on this. Know Sabeans' media personality a bit, I wouldn't be surprised if he thinks this is a major competitive advantage.

I have a feeling that having Bonds getting thousands of walks, often in weak lineups has colored his position on this. Know Sabeans' media personality a bit, I wouldn't be surprised if he thinks this is a major competitive advantage.

Are you saying that Sabean learned that Bonds getting walks was what lead to the weak lineups? And thus should be avoided?

I think he saw a situation where people would take the bat out of the hands of his best player - often his only good hitter (obviously, the Kent years were one of the exceptions).

And I think that while most analysis indicates that he was walked "too much" opposing managers kept walking him - hence it must have been a good strategy - hence walks are overrated. It's really more of a gut feeling on my part more than anything tangible.

GM Psychoanalysis - worst blog idea evah?

No worse than doing Kremlinology. :)

Clearly, Sabean doesn't know the value of the walk. It would be pretty amusing (in a tragic way, since I'm a Giants fan), that Sabean's exposure to one of the greatest batters of all time, who had no problem using the walk as a weapon reinforced his view that walks are not great outcomes.

This thread was hilarious.

It would be pretty amusing (in a tragic way, since I'm a Giants fan), that Sabean's exposure to one of the greatest batters of all time, who had no problem using the walk as a weapon reinforced his view that walks are not great outcomes.

Actually, it's not at all surprising. The Giants, for the most part, did a poor job of leveraging Bonds's walks into runs, and the lesson someone might take from that is that walks tend to be issued in situations where they are far less valuable than statistical analysis would have you believe. (Ripple effects from walks, e.g. extra PAs for your better hitters, are not often considered in that context.)

-- MWE

Looks like ZiPS is doing better than the Allied Runzler League so far.

Hey Dan Runzler! 19 BBs in 30 IP! He's still got it!