May 25, 2013

THT Essentials:
Fangraphs Player Search:


And here's the full roster.

Now available


You can now purchase the Hardball Times Baseball Annual 2013, with 300 pages of great content. It's also available on Amazon and Kindle. Read more about it here.

THT's latest e-book


Third Base: The Crossroads is THT's new e-book, available for $3.99 from the Kindle store. The good news is that anyone can read a Kindle book, even on a PC. So enjoy the best from THT in a new format.

Most Recent Comments





Get your very own THT merchandise from our CafePress store. We've got baseball caps, t-shirts, coffee mugs and even wall clocks with the classy THT logo prominently displayed. Also, check out the THT Bookstore. Please support your favorite baseball site by purchasing something today.



Or you can search by:


Creative Commons License
All content on this site (including text, graphs, and any other original works), unless otherwise noted, is licensed under a Creative Commons License.
Roll mouse over date for entries
THT Live Calendar
May 2013
S M T W T F S



1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31

Tuesday, January 26, 2010

Introducing the 5 Tool Analyzer

Posted by Kevin Dame
I've always been intrigued with the concept of the "5 tool player" and have been experimenting with ways to visualize what such a player looks like. The 5 Tool Analyzer maps a player's skill (relative to his peers) across 5 offensive and defensive measures. The resulting map reveals a few things at a glance. First, the overall size of the map indicates how good the player is. Big glob of color = good. Small glob = bad. Second, it shows where they are strong and where they are weak. And finally, it also shows how balanced a player is. Circular glob = well-rounded (literally and figuratively). Weird-shaped glob = some strengths and some weaknesses. Let's start with a quick visual introduction:

image


Now that we're clear on what we're measuring, let’s take the 5 Tool Analyzer for a spin. To start with, I was curious to see if I could find a player who was “perfectly average” in 2009 and I think I found him in Dave Murphy (he’s roughly at the 50th percentile across all 5 dimensions).

image


By now you’re probably wondering if there is a “perfect player” in baseball (a player who is the best across all 5 dimensions). Using the 5 Tool Analyzer, the perfect player would be represented by a perfect circle filling out to the 100th percentile. What I discovered was that most all-star players have at least one weak (or average) area. But there were a few players who came close to perfection, and you’ll be surprised to see the one who came closest:

image


Chone Figgins was an intriguing example of a player who truly excels in all but one category (in his case power):

image


Adam Dunn’s visual map paints a classic portrait of an “all bat” player. He has DH written all over him:

image


Jack Wilson represents the inverse of Dunn – an “all glove” player.

image

So that's a quick tour of the 5 Tool Analyzer. if you like it and want to see more players visualized, name your player!



"The commonality between science and art is in trying to see profoundly - to develop strategies of seeing and showing." - Edward Tufte.

Feel free to send comments, questions, and suggestions to Kevin via email.



Comments

John said...

Pujols

Posted 01/26  at  02:47 AM
200tang said...

Ichiro
Adam Jones
Matt Kemp
Elvis Andrus
Carlos Zambrano wink

Posted 01/26  at  03:14 AM
Conrad said...

Chase Utley

Posted 01/26  at  03:24 AM
Dan Novick said...

Kevin—

These are awesome.

Showdown between The Upton Brothers.

Posted 01/26  at  03:29 AM
Dick Smith said...

Vernon Wells.  Just want to see what 21 million dollars get you…..

Posted 01/26  at  03:39 AM
archilochusColubris said...

Hey Kevin, nice work, these graphics are by far my favorite of what you’ve posted here.

How bout someone like Lastings Milledge, who hasn’t fully developed yet?

Posted 01/26  at  06:31 AM
GSS said...

Jayson Werth

Posted 01/26  at  08:08 AM
John Walsh said...

Very nice.  A question: for the rate stats (bat, power, speed), who gets included in the percentile calculation?  Do you have some PA requirement?

Posted 01/26  at  08:45 AM
Andy Everett said...

I’d like to see these used to represent a players development over the seasons in 3D.  No small task I know!

Posted 01/26  at  09:19 AM
Dave Studeman said...

Wow.  I love the “blob/color” approach, Kevin.  I think this is the best example of this sort of graphic I’ve seen.

Having said that, I’ll raise the usual issue I have with these sorts of graphs: they graphically imply that all five dimensions are of equal worth, which is clearly not true. I don’t know of any easy way to fix this issue, but it’s important to remember.

Posted 01/26  at  09:30 AM
Ralph Kramden said...

While the results might not be as aesthetic, I think the analytical value of this visualization might be improved if the “arm” and “glove” scales were adjusted to better reflect their relative importance.  There is no position for which these are as important as bat and power.  And the “arm” component for 1B-men is much much less important than, say, C or SS or RF.

You’d wind up with ovoids instead of circles, but the overall size of the “blob” would better reflect the player’s value.

Posted 01/26  at  09:36 AM
Jamie said...

A-Rod
Jeter
Longoria

Posted 01/26  at  09:37 AM
Pat said...

Derrek Lee
Pujols
Jeter
Ryan Braun

Posted 01/26  at  09:48 AM
Nicholas Montgomery said...

Yuniesky Betancourt
Jeff Francoeur

Posted 01/26  at  09:57 AM
Tom M. Tango said...

wOBA already includes SLG.  You would be better off putting in OBP in its place.

I’ll also agree with the others that the five dimensions as presented makes it seem as if they are all equally valuable.

In order to do what you want, and keep the diameter the same for each dimension, you need to change the “slice” of each to not be equals.  For example, if the spread in SLG (in terms of runs) is twice that of the glove, then the slice of the SLG should be twice the area of the glove.

I can try to work out those numbers if you like.

Posted 01/26  at  10:21 AM
Nicolas said...

matt kemp
andre ethier
james loney
casey blake

Posted 01/26  at  10:43 AM
GWR said...

Dave had said: ‘they graphically imply that all five dimensions are of equal worth, which is clearly not true’

I am not sure why this is clearly untrue. At times a players arm may be as valuable as anthers speed or power. I think players having a big blob in one area of the graph shows their strengths and weaknesses and is useful for valuating that player. Just as this graph would indicate the value of Albert Pujols it would indicate the value of Jack Wilson. I think all 5 dimensions are of equal value.

Posted 01/26  at  11:53 AM
MikeS said...

So do we call players like Figgins “Pac Man” players?

Posted 01/26  at  11:55 AM
Ken said...

Are the percentiles position adjusted? ie. Kendry Morales might have the highest Double Play Runs Above Average for a first baseman, but I don’t really see him as having the arm of a 5-tool player. I’m not sure how you compare infield and outfield arms but it seems that some calculation is necessary.

Similarly, DPR is on average 0 for second-basemen and shortstops, but shortstops should get a higher rating for the Arm category - is it possible to adjust for that?

Posted 01/26  at  11:56 AM
kds said...

Following Studes, Jackie Gleason, and Tango, I would put league average at the 50% line and move in and out from there based on runs above or below average.  Whether to take absolute runs or adjust for playing time, (runs per 150 games, or 600 (650) plate appearences) is something I could go either way on.  It would make a difference with part time players such as Nyjer Morgan.

Posted 01/26  at  11:58 AM
Dave Studeman said...

I like Tango’s suggestion a lot.  I hadn’t thought of that before.

Posted 01/26  at  12:22 PM
Mitch said...

This is awesome.  I think “everyone” is my only request.

Posted 01/26  at  01:25 PM
Dwezilwoffa said...

These are absolutely fantastic.  Reminds me of the old Little League Baseball game ratings on NES.  And I second Mitch,  everyone will do nicely.

Posted 01/26  at  02:13 PM
Dave Studeman said...

Michael Bourn and Willy Taveras

Posted 01/26  at  03:08 PM
Bryz said...

I agree that these 5 tools are not equal in value, but I don’t think they need to be changed at all. These are merely ranking players in relation to the rest of the league.

Posted 01/26  at  03:55 PM
Bryz said...

By the way, how do you plan on judging arm and glove for catchers and 1st basemen?

Posted 01/26  at  04:39 PM
Ken said...

If you incorporate Tango’s suggestion of altering the size of the different elements, you should also convert all 5 tools to a Wins (or Runs) Above Replacement metric. Then the area inside the curve would have the interpretation of total WAR so that you would have a visual interpretation of total ability.

Posted 01/26  at  04:56 PM
Dave Studeman said...

I don’t see why you have to convert each metric to runs above/below replacement.  Tango’s idea to change the slice of the pie effectively does that for you.  No need to change anything else.

Posted 01/26  at  05:03 PM
liamandme said...

pedroia

Posted 01/26  at  07:06 PM
Ryan said...

Not adding much here, but damn those are some sexy graphs.

Posted 01/27  at  01:49 AM
Non Euclidian Fan said...

Any thought of replacing the circle with a fractal?

Posted 01/27  at  11:30 AM
Aryn said...

Those are quite beautiful.

The “growth” visualization via the evolution of the graph over time would also be quite a site.  As would a rendering of a lifetime “blob” in 3D.

And you forgot the “intangible” stat ... gotta have that in there.  A full text search of baseball “human interest” columns coordinating positive and negative press, with park factors of course ... for the inevitable “intangibly above average” stat!

Posted 01/27  at  01:32 PM
Craig Brown said...

Would Yuniesky Betancourt’s graph look like a black hole?  This is a great way to visualize the Vortex of Suck.

Posted 01/27  at  03:08 PM
Blake Atkins said...

Adam Jones
Nick Markakis
Brian Roberts

Posted 01/28  at  12:47 PM
Joe R said...

I was going to say it, but Tango beat me to the punch: wOBA includes power numbers, so wouldn’t OBP make more sense there? Adam Dunn’s chart for example probably wouldn’t change, but I can see this overrating a hacktastic power hitter like Miguel Olivo or Mark Reynolds, or someone else whose wOBA is heavily dependent on his power (since he has twice the area to fill up)

Posted 01/28  at  02:20 PM
Matt Keller said...

Robinson Cano, Bengie Molina

Posted 01/28  at  09:14 PM
Phylan said...

I’m actually not really surprised that Zobrist was closest to perfect, to be honest.

Posted 01/28  at  09:21 PM
Dan Novick said...

Tango, I think you’re wrong here with the suggestion to use OBP instead of wOBA. Walks are not one of the five tools. As much as it may hurt to say it, I think batting average works best here.

You can’t use OBP in looking at the “bat” tool because such a big part of OBP is walks. What’s OBP minus BB? Batting average.

Posted 01/28  at  09:22 PM
Joe R said...

Dan, maybe BB isn’t one of the “traditional” tools, but these days, most GMs and scouts value both the ability to hit for average AND walk. So I think OBP > BA for this exercise.

Posted 01/29  at  10:33 AM
Dave Studeman said...

Tango, I think you’re wrong here with the suggestion to use OBP instead of wOBA. Walks are not one of the five tools.

But they clearly should be.

Posted 01/29  at  01:37 PM
Web said...

First let me say I absolutely love your idea.

I would like to see comparisons by position and league.

Pujols, Gonzalez, Fielder, Lee, Helton, Berkman, Howard, Cantu, etc… for the NL for example.

And I really like the idea of weighting the graphic by size of wedge as Tango was talking about. That would give a better representation of overall talent.

Thanks!

Posted 01/29  at  04:35 PM
Heineken-77 said...

JOE MAUER!!!
Justin Morneau!!!

Posted 02/03  at  11:48 AM
Page 1 of 1

Leave a comment:

Commenting is not available in this weblog entry.