Marcel projections are out.

[RedbirdRay]

marcel

 

tangotiger put up the marcel projections a few days before xmas. didn't think they were posted here yet. Either my formula for slugging is messed up (they are not calculated on the spreadsheet, you have to figure it yourself) or there are several guys projected above .700. probably my least favorite of the projections anyway.

[rluzinski]

I was going to post this on my blog but didn't get around to it. Here are the Marcel projections for Brewer hitters in 2007:

 [b] BA OBP SLG OPS[/b] Jenkins .275 .351 .465 .816 Fielder .285 .357 .493 .850 Weeks .274 .359 .426 .784 Hardy .264 .337 .424 .760 Koskie .256 .338 .450 .788 Estrada .289 .336 .422 .757 Miller .260 .329 .404 .732 Hall .276 .338 .498 .836 Hart .277 .336 .462 .798 Gross .265 .353 .433 .786 Counsell .252 .333 .362 .695 Graffanino .278 .344 .401 .745 Clark .282 .359 .399 .758

This is one of my favorite projection systems since it's the simplest and gives up only a small amount of accuracy to do so. These numbers aren't park adjusted, so projections from a guy like Estrada (who's been playing in a hitter's pack) will be inflated).

 

Ray, when given h, d, t, and hr, the SLG equation should be:

 

SLG = [h + d + (2 x t) + (3 x HR)]/AB

 

EDIT: Here's the methodolgy behind these projections:

 

LINK

[RobertR]

Those projections can inspire some optimism. I'm convinced that the most important thing for the Brewers is good health. Give me relatively good health and those projections, and I'll say the Brewers are likely contenders if not favorites.

 

Robert

[RedbirdRay]

Ya, I'm so dumb. I was counting total hits as singles on my spreadsheet. Trying to be too fast and not thinking about what I was doing as usual.

 

I really respect the simplicity of it. I guess it's my least favorite due to it's ranking with the other systems. Seems most do at least one thing pretty well (either pitching or hitting) and marcel is bottom of the pack on both.

[Klements Bratwurst]

Does anyone have data/link to what the average OPS is per position by league? I am curious to see how these projections for the Brewers compare to what the league average is per position.

[RobertR]

You can find positional averages here.

 

Robert

[thebruce44]

Wow. The average NL OPS for 1B was .879 compared to only .819 for the AL. I am guessing thats thanks to Howard/Pujols.

[sbrylski]

So these projections would penalize things like Estrada playing through injuries in '05, correct? I know they are not perfect, but fun to look at.

[Klements Bratwurst]

Ok here goes a quick analysis of the Brewers starters and nothing scientific. Thanks Robert R for the link.

 

MLB average is a simple average of both leagues.

 

Brewers Averages assume a simple average for catcher of Miller and Estrada OPS. Geoff Jenkins remains the rightfielder. Corey Hart is the leftfielder. Hall in Center.

Koskie- 3B, Hardy SS, Weeks, 2B, Fielder 1B

 

 

 

Pos AL AVG NL AVG MLB AVG MIL AVG

C        749        743        746        744

1B        819        879        849        850

2B        727        758        743        784

3B        780        826        803        788

SS        745        737        741        760

LF        796        837        817        798

CF        770        753        762        836

RF        817        798        808        816

[1992casey]

That Baseball Prospectus positional average page is much tougher to read than its old text-only one. http://forum.brewerfan.net/images/smilies/frown.gif

 

I assume position 10 is DH and position 11 is PH. What are positions -1 and 12? (Could 12 be pinch runner?)

 

Klements, it looks like you took each NL positional average OPS and averaged it with each AL positional average OPS. That's probably good enough for a 'quick and dirty' calculation.

 

But note that the NL has more teams, and thus more plate appearances. To be accurate, weighing each average according to plate appearances would probably the be the best.

[dadofandrew]

12 has to be PR, based on the SB's and runs scored.

 

-1 happened so rarely, but much more in the AL...P having to bat because the DH had to play the field? No, I don't know why this would not go under P.

[Beast Light]

I think that it is interesting that the pitching projections total up to 3,291 wins and 3,290 losses.

 

I wonder how this is supposed to happen when there are only 2,430 games played per year.

 

I know that they are only projections but I put zero stock in them. When I see things like Prince being projected to hit 22 home runs it makes me think that the logic is not credible.

[filthyfrog]

Wow. The average NL OPS for 1B was .879 compared to only .819 for the AL. I am guessing thats thanks to Howard/Pujols.

 

Also, some of the best would-be 1Bs in the AL are DHs.

 

While Klements' numbers look pretty exciting, you have to remember that the Brewer projections don't take into account the fact that many at bats will be from back ups and replacements. Even if all the starters stay healthy, guys like Counsell and Clark are going to play occasionally, which will bring down the overall numbers. My guess is that most teams would look pretty good if you compared their projected starters to positional averages.

[Klements Bratwurst]

Quote:

---

But note that the NL has more teams, and thus more plate appearances. To be accurate, weighing each average according to plate appearances would probably the be the best.

---

 

How could I forget something so simple regarding the NL having more teams? I just opted for the quick and dirty approach for sure, just to get an idea. Filthy Frog has some good thoughts too in relations to reserve players. I suppose you could plug in Gross somewhere, Clark somewhere, but at how much at each position?

 

However, I do like Hall in Center though!!! Or SS for that matter!!!

[DHonks]

i'm curious why they have Koskie's OBP so low, and Weeks' slg the same as Hardy.

[dadofandrew]

Rickie's career SLG is .397, JJ's is .388...they should be projected close.

[sbrylski]

I know that they are only projections but I put zero stock in them. When I see things like Prince being projected to hit 22 home runs it makes me think that the logic is not credible.

 

That's because these are statistical analysis's done in replacement for scouting. Scouting rules the roost in this area of projections, especially young players. I'm guessing Fielder's HR projection is lower because it is simply comparing what he did last year to what a normal 22 year old 1B does and averaging them. (That was an extreme simplification.) Nowhere does it take into account potential, work ethic, ect. Fielder should be compared to how the other top ranked 1B prospects did in their second season and projecting their improvement onto him, IMO.

 

Note: I may be quite offbase here, as I know little about the Marcel projections and didn't really go into any of the links.

[1992casey]

---

i'm curious why they have Koskie's OBP so low

---

I'd guess that Corey's getting hit by his poor .735 OPS campaign in 2005. Toss in that he was 32 that season and the .788 OPS projection fits right in.

 

---

I know that they are only projections but I put zero stock in them. When I see things like Prince being projected to hit 22 home runs it makes me think that the logic is not credible.

---

I wouldn't knock the credibility all the way down to zero. Rather, I'd say "know thy projection system" and go from there. These things are tools and tools must be used properly. Keep in mind that by necessity, a computer must use a 'one size fits all' approach. If it didn't, it'd stop being objective. If I'm a GM, I'd want to use these systems to keep my feet on the ground, so to speak, when trying to do my own more subjective projections.

 

Counting stats, such as home runs, will be affected by projected ABs. Prince's 2005 AB total amounted to less than that of a full-time player. While league effects should be accounted for, Prince only hit 23 homers in 2004 at Huntsville.

 

In short, younger players are often projected to have a goofy number of ABs and counting stats are affected accordingly. Even the rate stats (averages) need to be handled with care. If you know a guy's likely to get 250 ABs and he's projected for 450, you'll have to give the numbers some breathing room. Slumps or hot streaks are much more likely to stick out.

 

You might want to take a look at the Projection systems thread Ennder started a few days ago; it contains a couple of links that might be helpful. (I can't attest to how helpful they might be; I haven't read them yet. http://forum.brewerfan.net/images/smilies/wink.gif)

[dadofandrew]

This is why I don't understand the indifference about how improved the Brewers' depth is. A falloff from Hardy to Counsell is about a game, using the 10 run rule, and Weeks to Graffy even less than a game. Bruan might be better than Koskie, or at least neutral if you figure in his supposed defensive issues, which may or may not be real.

 

Pretending there is some huge disparity between JJ and Craig is just incorrect and emotional. If you think the Crew will win 88 games with JJ as the SS, the math puts them at about 86.8 with Craig. That is not much falloff at all, and oodles better than most.

 

The one guy this team cannot afford to lose for a significant period is Fielder. A solid AAA vet at 1B should be on Doug's "to-do" list.

[RobertR]

I might put Sheets or Cordero on the more indispensable list than Fielder, but I agree that Fielder would be the toughest of the position players to replace, maybe with Hall in CF being the other spot where the Brewers would take the biggest hit. That said, in my mind, Fielder's backup is Braun. And I feel a lot better about the Brewers with that option at AAA to backup 3B, 1B, or corner outfield, if necessary.

 

Robert

[RedbirdRay]

Marginal wins don't seem to mean much, but they do add up. The difference between 82 and 84 wins might not be a big deal, but the difference between 91 and 93 wins is huge.

[SoCalBrewfan]

(correct me if I'm wrong here) Marcel doesn't use minor league stats, so Prince's numbers would be based on his major league track record. His 2005 numbers are of little use for projection, but are included anyway IIRC. To be sure, that's one of the limitations of the system, which explicitly doesn't try to be all things for all players.

 

For Prince in particular I would probably take the 'over' on his Marcel slug%...ZiPS is another system that does use minor league numbers, and projects .280 / .358 / .513 with 30 homers. Note also that the Marcel projects only 530 plate appearances, while Fielder actually had almost 650 last year. With his '06 plate appearances he'd have a projected total of 28 HR...which makes some sense because the system only has '06 and a small sliver of '05 with which to calculate.

 

The nice thing about the Brewers is that they have a number of young guys, and it's very possible that one or more will take large steps forward. The likely candidates for decline and/or injuries are at positions where there's some depth (C, 3B, 2b/ss, LF, SP), though multiple injuries and bad years of course would be tough to overcome.

[1992casey]

---

The one guy this team cannot afford to lose for a significant period is Fielder.

---

I think I'd put Hall in that category just because he's Billy. I also might be tempted to include Hart as a runner-up simply because he's a guy that can be depended on to hit both lefties and righties.

 

---

That said, in my mind, Fielder's backup is Braun.

---

That's an interesting thought. I think that if that happened, the team would decide to let Ryan continue his development at 3B and move Koskie to 1B. That wouldn't be a good move for the team's defense, but it'd probably be a good move for Ryan.

[rluzinski]

Yes, Marcels use only MLB stats, so projections of the young guys are pretty bare bones. If you download the full spreadsheet, it includes a reliability column which is basically tied to sample size. He's basically assuming Prince will only get better by a league average amount a player like him would improve. Prince is not a league average player, however.

 

The more AB a player has at the MLB level, the more his performance is likely to represent his actual abilities. I like marcels because it's completely objective. Since we know more about these guys than Marcels does, we can adjust accordingly.

 

Also, Counsell's projections are based primarily from stats he accumulated at a hitter's park. His projected numbers at Miller Park would be lower. Same for Estrada.

[RobertR]

That's an interesting thought. I think that if that happened, the team would decide to let Ryan continue his development at 3B and move Koskie to 1B. That wouldn't be a good move for the team's defense, but it'd probably be a good move for Ryan.

 

Well I was speaking mainly in regards to replacing Fielder's bat. I'd expect the Brewers would lose something in defense if they moved Koskie to first, but I'd also suspect that the upgrade from Fielder to Koskie would mitigate a great deal of the loss.

 

I probably should rethink the scenario though. If Fielder would go down early in the year, I suspect the substitution would be Hart to 1B and Gross/Clark to a corner outfield position. Later in the year Braun is going to come into the mix. While neither scenario is ideal, I think the Brewers could muddle through a Fielder injury. I'm still thinking that a Sheets or Cordero injury could be much more devastating to the Brewers. Those are their All Star caliber players.

 

Robert