Omar don't let me down, claim him!

Wow! I wasn't expecting this.

Reportedly, nobody put in a waiver claim on Todd Walker (different waivers) and I think all the contenders are pretty set at 2B, except I guess the Mets, so Bellhorn may make it through. The Cards grabbing him for 3B might make sense but he doesn't seem a LaRussa player and I think they're fine with Nunez anyway.

Walt,

You must be mistaken. The Mets have Miguel Cairo at second-base. Obviously, they are set. I mean, tell me, if the Mets were to replace a key hitter like Miguel, who would replace him in the 2 spot? Who else could bunt Reyes to third and put the ball in play? Who?

Let's see it Mr. Terry Ryan..I'm as big of Brent Abernathy fan as the next guy but really he's Brent Abernathy.

I expect Bellhorn to pass through waivers. The question is then whether he'll declare free agency or accept the assignment to Pawtucket. If he can get a starting job elsewhere (KC?), I want him to declare free agency. If he's just going to be a back-up, then I want him to stay in Pawtucket.

SIGN HIM METS!

The question is then whether he'll declare free agency or accept the assignment to Pawtucket.

He's now behind Graffanino, Cora, and probably Pedroia, and I'm sure he realizes that.

I'll be shocked if he agrees to go to AAA; he gets paid his contract whatever he does, so $$$ won't be an issue.

Yeah, im sure he will declare FA, their has to be a team out their that will give him the starting job

SportTicker: Red Sox designate 2B Bellhorn for assignment

The Boston Red Sox parted ways with one of the heroes of their 2004 World Series Championship team on Friday, designating second baseman Mark Bellhorn for assignment.

Bellhorn had been on the 15-day disabled list since July 18 with a sprained left thumb. He struggled offensively prior to the injury, batting just .216 with seven home runs and 28 RBI in 85 games.

Bellhorn failed to make much progress in his rehab assignment with Class AAA Pawtucket, batting just .176 (12-for-68) with two homers and 24 strikeouts in 68 at-bats.

NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

I thought when Boston picked him up they were demonstrating that they could see past his annual slow starts, huge K totals, and incosistency, and instead focus on the fact that when he gets consistent PT, he ends up with good numbers.

But that is apparently not the case, so much, anymore. Bellhorn will be an assett to someone, as an average 2B with a + bat. Hopefully, it’ll be for the Red Sox.

Look, Bellhorn is (was? *sigh*) my favorite player on the Red Sox, and I'm sad that he's going, but the above two paragraphs are silly. He'd been getting consistent playing time all this year and had still hit poorly. Saying that the Red Sox are apparently not able to see past the negatives for the fact that he has an above-average bat (given consistent playing time) is false -- he had plenty of playing time this year and still never got things together. Do I expect him to hit better next year? Of course, but in the meantime, there's no spot for him on the major league team. The only shame is that Petagine hasn't gotten the same sort of chance to replace a struggling player.

Yeah, I basically agree with Robert here.

Graffanino is a solid 10-run upgrade defensively, projects similarly as a hitter, and is better right now - if there's anything subjectively wrong with Bellhorn.

The Red Sox are doing right by him to offer him this release, and I hope he catches on and hits well for some National League team. I'll be rooting for Rumblefish wherever he goes. Really good guy, fun player, GAME 6!!11!!

Yes, nice post, Robert M.

He might have been my favorite player too. He has an Achilles heel - his bat control is poor (and by that, I mean he has trouble getting his bat on pitches). That is usually a fatal flaw but Bellhorn was able to fashion not just a career but some truly excellent seasons working around himself. He is a very intelligent, introspective player and I will miss him.

Best wishes, Mark. I'll be rooting for you to turn it around.

(Postscript- this may sound naive but by the way he just swings right through very hittable pitches, it seems to me he has some visual problem, like a depth perception thing or something. He should get his eyes reexamined and then consider opening his stance so both eyes are fully focused on the approaching pitch).

So, just maybe, the Cubs weren't complete idiots for releasing him after all.

No, they were still idiots. They missed a year of this:

.264 .373 .444

57 X-base hits and 93 runs scored is nothing to sneeze at from your 2-sacker.

There has to be a team that could use a Mark Bellhorn. Of course, that team may not realize that.

See my post on the other thread. Essesntially, despite his poor start (which is the norm for him) and Graff's good one, he's still projected to outhit Graff. He's certainly projected to outhit Cora. He's struggled during his rehab, but it is rehab afterall. I believe he was okay in his last couple games:

2-5 2B
2-4 BB
1-5 2B
1-5 2B
1-4 HR

Total: 7-23, 3 2B, 1 HR, 1 BB, .304/.333/.565.

(Normally, I don't like to pull Eric Van's like the above, but in this case, we're talking about someone rounding into shape after an injury. The most recent 5 games seem to me to be most relevant.)

I'll always remember the home run he hit in game one of hte world series on my birthday!

I don't like to pull Eric Vans like that either. Because if you can do it for Bellhorn, you should do it for Graffanino. His last 7 days (or 4 games), according to ESPN: 7-16, 2 2B, .438/.438/.563.

Again, believe me, I think Bellhorn is likely to regress to the mean (i.e., I think he will hit better in the future), but in the meantime (no pun intended), Graffanino is outhitting him now, has outhit him all year, and has generally been a decent hitter (with lower highs than Bellhorn, but higher lows) and a good, solid fielder (like Bellhorn). I just don't see a problem with choosing Graffanino over Bellhorn at this point. Could it work out that Bellhorn returns to how he hit in 2004 and 2002? Sure. It could also work out that he hits like he did in 2005 and 2003.

Again, the real irritant is that Bellhorn gets DFA'd and Millar merely gets platooned when Bellhorn was better, relative to his position, than Millar. (I think that's true on just offense, but it is almost certainly true when defense is figured in). It feels like a victory of extroversion over introversion, when I'd rather there be a victory through simply playing well (or, in Bellhorn's and Millar's unfortunate cases, playing better).

dammit, I hate this. He had become my favorite player.

I don't like to pull Eric Vans like that either. Because if you can do it for Bellhorn, you should do it for Graffanino. His last 7 days (or 4 games), according to ESPN: 7-16, 2 2B, .438/.438/.563.

I think it's obvious that I was only using that small amount of data becuse Bellhorn had been hurt. It was only meant to point out that Bellhorn's putrid stats in AAA, are consistent with a guy rounding back into shape after an injury.

Bellhorn would also be a better option than Cano right now.

Just moving this over from the other thread:

Coming into the year, ZIPS had:

Bellhorn .267.384 .450
Graff .253 .325 .352

Including this year's production, Dan now has Bellhorn slated for this:

.246 .364 .413

Not sure what it says about Graf, but do you think he's likely to do better than that?

Well Dan Szymborski's computer said Bellhorn *might* be a better hitter this year, so he should play?

Eh, RM said just about everything I wanted to in #20 anyway. It's a shame to see him go, but so it is.

Well Dan Szymborski's computer said Bellhorn *might* be a better hitter this year, so he should play?

Well, it said he *would* be better. Are there other projections that disagree with this? Did you think Graff would be better coming into the year?

Well, it said he *would* be better. Are there other projections that disagree with this? Did you think Graff would be better coming into the year?

Shandler had Graffinino at 262/328/378 and Bellhorn at 249/324/432. Some of Bellhorn's 50 pts SLG edge is probably just Fenway.

Last year UZR had Bellhorn at -5. Graffinino has always been a UZR star right? So Shandler + UZR would have projected Graff as a better player this year.

Last year UZR had Bellhorn at -5. Graffinino has always been a UZR star right?

Where are people getting 2004 UZR numbers? Bellhorn was better in 03 and I think Graff had the edge in 02. Probably give an edge to Graff overall. That would make him a better player than Bellhorn.

Whether he's better than Graff or not, I think there is a place for Bellhorn on this team, ahead of Cora and Millar.

Whether he's better than Graff or not, I think there is a place for Bellhorn on this team, ahead of Cora and Millar.

ok i agree with that. but i'd add, that he's behind youks, petagine, and petunia in the 25 man roster in my mind.

Where are people getting 2004 UZR numbers? Bellhorn was better in 03 and I think Graff had the edge in 02. Probably give an edge to Graff overall. That would make him a better player than Bellhorn.

MGL posted some random ones at some point.

Bellhorn was better than Graff in UZR at 2B in 2003? That doesn't make much sense as Bellhorn only played 118 innings at 2B in 2003. Whatever his UZR number was for 118 innings the sample is way too small for it to have any meaning.

In fact, the same can be said for all of Bellhorn's 2B UZR. In the period that MGL studied and posted about - 2000-2003 - Bellhorn only played 777 innings at 2B. Tango has siad you need about two full seasons of defensive play to be as confident in the UZR numbers as you'd be in one season of hitting numbers.

So you can think of the meaningfulness of a half season of UZR as similar to a quarter season of AB, which of course no one would do.

That Bellhorn is a good UZR defender is a myth based on small sample sizes.

ok i agree with that. but i'd add, that he's behind youks, petagine, and petunia in the 25 man roster in my mind.

Being as he was already on the 40, I'd say that also puts him ahead of Pedroia.

So you can think of the meaningfulness of a half season of UZR as similar to a quarter season of AB, which of course no one would do.

I'm not sure what you mean by this.

That Bellhorn is a good UZR defender is a myth based on small sample sizes.

I had never considered this. It's a good point, though I haven't been saying that he's a + defender. During the same period (00-03), Graffanino had 672.7 innings at 2B. So if we're comparing Bellhorn and him, we should probably throw out both of their numbers.

In 2004, Bellhorn racked up another 1044 while Graff got another 630. Both of those samples alone would not be meaningful by the Tango standard. But when combined with their other numbers, would make for a reasonably meaningful sample, right? For Bellhorn, that's 777 innings at +17 and 1044 at -5, so he'd be +4 or so. For Graff, it'd be 672 at +21 and 630 at ???. (Even then, you've got a considerably smaller sample for Graff)

One other thing to consider is their defense at other positions. Graffanino grades out high at SS and 3B too, which makes me more confident that his 2B numbers are for real. Bellhorn gets good grades at 3B, but poor ones in a scant few games at SS and 1B. I'd say these also serve to legitimize Bellhorn's 2B rating.

On a side note, Cora got a -6 at 2B (142 "games") and +12 at SS (248) in the 00-03 period.

Farewell, Mark. The "real" MVP, offensively, of the 2004 WS. Sniff.

Yeah, I wish Bellhorn had gotten the WS MVP that he deserved. Then maybe the fans wouldn't have shat all over him quite as much.

philly is correct about what I said. The hitting stats of 600 PA by a hitter has the same certainty level as the UZR of 1200 balls in play opps for a fielder. For 2b/ss/3b/cf, that'd be about a season and a half to 2 seasons, and for the 1b/lf/rf, that'd be 2 1/2 to 3 seasons.

Since we'd like to look at 3 seasons of data for a hitter to feel really comfortable, you'd like 6 seasons of UZR to have the same comfort level.

Actually, to be more precise, this is true only if a person does not age. So, 200 PA and 400 BIP, since they happen in the same year, a person will not have aged much, and so, we'd have the same certainty level.

However, once you get into multi-years, you really need to also consider the aging factor. Once you do that, 2 years of hitting is equivalent to 6 years of UZR. 3 years of hitting is equivalent to an entire career of UZR.

However, if you were to also include the Fans' Scouting Report, 3 years of hitting would be as certain as 6 years of UZR+Fans.

Tango,

That is to say, that when you regress UZR 2006 on UZRpast and FSR you improve the R^2 significantly?

Have you compared the FSR to, for example, BA's scouting numbers, etc.?

IIRC, the r goes from .35 (Fans only) to .50 (UZR only) to .60 (UZR + Fans), when compared to next year's UZR. That the r-squared add up so neatly (.12, .25, .36) shows the independence between the measures.

No, I have not compared to BA's scouting numbers, but if Kevin or Alan or someone there wants to provide me with electronic data, I'm sure we'd get some fascinating results.

My guess is that 20 hardcore fans = 2 professional scouts. Since I've got around 700 ballots, I probably have the equivalent of 70 team scouts. If BA has a scouting staff of 5 guys each assigned to 15 teams, we probably have the same reliability. Just guessing.

Tango,

The combination variable UZR*FSR didn't do much extra explanation? I wonder if components would help explain more. For example, UZR*Speed for CF vs. UZR*agility or something for 2B.

Also, I think that sounds about right with the scouts, though maybe The Wisdom of Crowds just means that a group of fans is just better. It would matter what scouts, what fans, etc. but I'm not so sure that 2 professional scouts could ever be as good as 700 fans.

Actually, the UZR*FSR did as much as could possibly have been expected. Remember, the r-squared of the fans was .12. And UZR was .25. So, the absolute most you could have hoped for was .37, and we got .36.

The Fans' positional run values already weight each of the 7 traits differently for each position. If I were to only look at the Speed for CF, I could not get an r-squared of .12. The combination of speed, plus the other 6, gets me to .12.

In fact, for CF in particular, the r-squared for speed might not even be high, since there's not that much variance in speed among all CF (i.e., they are all fast). This is similar to what I did for 3B and arm strength (see my site, go down to the Scouting Report section, and click on the 3B report).

Tango, I wonder if, by way of the fact that they get to see it constantly, scouts can tell us a lot more about certain factors. As fans, we watch mostly on TV, and see speed, arm, and hands very easily. We also see how well the DP gets turned, and have an OK idea of the jump a CF gets on a ball because we can see how hard they have to move back, if they seem to be under it, etc.

We see very little for jump from infielders, or first step. We also get very little info on footwork, etc. My guess is that that is where the scouts would come in handy.

Also, the R^2 for Marcels on hitting is about .68 or so, right?

FWIW, as I've tried to train myself to watch ballplayers, I've found that footwork is pretty easy to see, when you watch for it.

Regarding defense stuff generally, MGL posted in the main board Rumblefish thread that Renteria's UZR this year is +2.

This generally comports with my analysis of the Sox fielding - where they should have been good, they've been average, where they shoulda been average, they've been below, and where they should be bad, they've been, yup, bad. I guess it's what you expect from an older team - defense peaks the moment a player enters the big leagues and declines from there. But it's part of what makes this such a frustrating team to watch.

tbfg always says, "TGWWTWSLY" - thank god we won the world series last year.

I'd add, TGWHTWBLA(EWF) - thank god we have the world's best lineup again (even without freedom).

Joel, I agree with you.

***

I don't remember if that was r or r-squared. I figured out once that the absolute highest you can get it to (using only PA-level data, and not pitch-level data) is .75 (same scale, can't remember r or r-squared).

Tango, maybe I should just STFU and do it myself, but have you run the regressions w/ say, Pinto, FSR, UZR, BPro (I hate the numbers, but maybe they tell us something). I know there will be a decent amount of multi-collinearity, but if all of these are picking up on something, it could be interesting to see how much explanatory power we could get from all the methods. Since finding a consistent method might be hard, one could always regress the sum of these numbers for each team in a given season, on the future season's PADE or something like that.