Johjima Community Projection Results

by Jeff on Feb 6, 2007 8:39 PM PST in Regular News

The spreadsheet's still open for edit, but updates have slowed to a snail's pace, meaning pretty much everyone who wants to participate (140+ people so far) has already done so. Making this a good time to share the mean results:

AB: 493
H: 143
2B: 26
3B: 1
HR: 20
BB: 31
HBP: 9
K: 51
BA: .291 (.017)
OBP: .344 (.019)
SLG: .468 (.032)
IsoPo: .177 (.020)
IsoPa: .053 (.010)

For those of you who're more visually inclined:

(The little bars above each column correspond to the sample standard deviation, with values of .025, .027, and .037, respectively.)

As an average, the community thinks Kenji's in line for a slight improvement going forward. While the batting average stays the same, the forecast calls for a 57% increase in walk rate and an 11% isolated power boost. While that might be unusual for most 31 year old catchers, Kenji is still adapting to the league, so if he begins to show more power in 2007 - which isn't exactly unprecedented - pitchers could very easily start working around him a little bit more, leading to more free passes. I imagine as far as the walk rate is concerned, there's also a little subconscious regression to the mean taking place, since his was one of the lowest in the league last year.

Kenji Johjima is a really good catcher, and if it weren't for Safeco, a lot more people would know it. If he gets any better he's a borderline star, and even if he stays the same or gets a little worse going forward, he's still a great deal at the price. Look elsewhere for your 2007 Collapse Potential.

Up next: Richie Sexson. Spreadsheet should be open and available by late morning tomorrow.

0 recs | Comment 10 comments

Story-email Email Printer Print

Comments

Can you explain to me deviation?

I understand that it represents some sort of maximum and minimum but I don't fully understand it.

Hot potato is an entirely different game when you're starving

by Mariner John on Feb 6, 2007 9:41 PM PST 0 recs

Oh man.

It's almost impossible for me to give a short answer to that. And naturally, Wiki should be your first stop.

Here's the easiest explanation I can give: the standard deviation is a measure of the variability in the data, and meant to be read as a +/-. So, if a data set has an average of 5 and a standard deviation of 1, the majority of the values lie between 4 and 6 (5 +/- 1).

As far as the Community Projections are concerned, a small standard deviation value means that people generally agree on what a player's going to do, whereas a larger value means there's a lot more disagreement.

by Jeff on Feb 6, 2007 9:50 PM PST up 0 recs

So...

Is it in anyway like regression where -1 < r < 1? The smaller the deviation is, the closer the regression would be to -1 or 1 and the larger it is, the closer it would be to 0?

Or more simply, deviation eliminates outliers for giving a range?

Hot potato is an entirely different game when you're starving

by Mariner John on Feb 6, 2007 10:14 PM PST up 0 recs

No, not really.

It is like r in that it shows you the agreement in the data set, but that's about it. It's a measure of certainty; in the case of community projections, the larger the standard deviation, the more unsure people are about where the player will end up.

by Jeff on Feb 7, 2007 5:04 AM PST up 0 recs

Standard deviatio

Is the spread in the distribution of the data set. In Jeffs example it's showing how much agreement or disagreement there is in the projections.

by chrisisasavage on Feb 6, 2007 9:58 PM PST up 0 recs

In this example

There is a lot more agreement on his projected BA than his OBP or especially SLG. However, we'd need to wait until all the projections come in before making conclusions about what that means.

by chrisisasavage on Feb 6, 2007 10:00 PM PST up 0 recs

As a group

were not to far off from most projection systems except for the power increase were perdicting.

How in the name of Zeus's butthole did you get out of your cell?

by Scruffy Lefty on Feb 6, 2007 10:04 PM PST 0 recs

which is likely a result

of the projections being out and influencing our thinking. Or maybe what's more at play here is that the projection systems arrive formulaically at the same conclusions we intuit as fans.

I wonder if our projections would be very different if this was run before Zips, Marcel, etc came out. Anyhow, Joh isn't the most interesting forecast. I think there will be a lot more disagreement between us and the systems when it comes to Beltre, Lopez, the pitchers, and so on. Guillen poses a nice problem as well (did all the tethered running pay off?).

by Mere Tantalisers on Feb 7, 2007 6:19 AM PST up 0 recs