I'm just not much of a stats guy

by Brad Wells on Jan 2, 2008 10:57 AM EST in Colts Analysis

I'm not a "stats" guy. Some people get really jazzed about things like DVOA, adjusted PPG, YAC, and other abbreviations which (in my not-so-humble-opinion) have little to do with what actually happens on a football field. As silly as it sounds, the only stats I look at are Ws and Ls. Unlike most other sports, professional football is a game of intangibles. Yes, things like YPP, Completion %, Points Scored, and TD to INT ratio are important stats, but even more important are intangible factoids like the competence of the Head Coach, the toughness of your QB, and the conditioning of the team physically (just to name a few). I know many here, and at other blogs, love the work Football Outsiders does. Their stat collecting is excellent, and they do have a good system for calibrating stats.

However, like most stats, the numbers simply don't add up when you see the final product on the field.

Case in point: Last season, Football Outsiders' DVOA ranked the Baltimore Ravens #1 over all other NFL teams. San Diego was #2 and the Philadelphia Eagles ranked #3. Meanwhile, the four teams that played in the Championship Games (Indianapolis, New England, Chicago, and New Orleans) ranked #7, #5, #4, and #9 respectively. The #1 ranked team (Baltimore) was physically dominated by Indianapolis, and San Diego was beaten at home by New England. New Orleans's 9th overall ranking meant nothing to them as they beat Philly, the best ranked team in the NFC.

The obvious point here: DVOA's mean absolutely nothing in the post-season, and I'm tired of people using them as a barometer for who is "the best."

Again, nothing against people who love stats (because stats are fun), but realistically stats don't tell you much about how good or great a team is. Last season, the Colts had the worst rushing defense in NFL history. They surrendered 173 rushing yards a game, but still won 12 games. For all of us here, we all knew that if Bob Sanders was healthy, if Cato June could re-learn how to tackle, if Gilbert Gardner was booted from the starting SAM position that those 173 yards were not a true reflection of the team. And look what happened: The Colts run defense dominated the 2006 season playoffs.

So, when people throw around numbers and rankings and all that stuff, please don't be offended if I pay them no mind. I'll remind you of what one person at Football Outsiders said prior to the start of this season; it's yet another reminder that Michael David Smith simply doesn't know anything about the sport in which he covers:

I'm the last guy to get caught up in the importance of skill position players over linemen, defense and special teams, but at the same time, I just believe a team with Peyton Manning, Marvin Harrison, Reggie Wayne and Joseph Addai is going to be really good, even if the rest of the roster is a bunch of guys who would be on the practice squad of other teams.

Again, Smith is a "stats guy" and likes to use his stats to back up his claims. After watching the depth of this Colts club really carry this team through an injury-riddled season, it's clear to me that Smith should spend more time watching games than reading stat lines. He might learn a thing or two, and not look like a total idiot.

Tweet Comment 8 comments | 0 recs |

Do you like this story?

Comments

To add to this

often just the overall number is listed, so while Chicago might have been #4, that makes them sound like an all around team. It didn't show the potential of our QB. It was just as likely that he threw for 3 TDs and 0 INTs and he was to throw 0 TDs and 4 INTs, so while our defensive and special teams numbers brought us up (we were one of the highest scoring teams last year), it doesn't properly represent a team.

It also doesn't show how a team reacts to a certain player. Your team's D didn't just get better because of Sanders being on the field and him being a good player. It got better just as much because the rest of your D reacts to Sanders and play better individually. If you best guy plays hard, you play hard. If he isn't around, you might not play as hard.

by WCG @ Stampede Blue on Jan 2, 2008 11:55 AM EST reply actions

Stats cause misrepresentation in all sports...

Being a DBacks fan, I'm very proud of this.
In baseball there is a stat much like DVOA, it is called Pythagoras. It is supposed to rank a team by runs scored, runs against, so on and so forth. Well the Diamondbacks after the stats were all said and done. AZ was supposed to finish 8 games under .500, AZ finished 18 games over .500. So, stats don't necessarily tell all. They are pretty good at gauging probabilities they don't tell the whole story. That being said I still love stats that tell the side of the story I want ;-)

by bleedingblue @ Stampede Blue on Jan 2, 2008 1:52 PM EST reply actions

I <3 Stats

I'm probably declaring my side in this argument just by saying I read and liked it but, Moneyball talks about where and why stats fail. When they talk about "Why Billy Beane's shit doesn't work in the playoffs."

Luck is huge. In the playoffs the teams are much closer in talent than in the regular season, so the best team will win less often. Are the Chargers better than the Colts or was the 6 INT game from Manning pretty fluky? The best team doesn't always win. If they did the worst team would go 0-16 and the best 16-0. The example in Moneyball is in the baseball playoffs the average difference in talent is about one run per game. The difference due to luck, four runs. The team that is best by stats isn't guaranteed to win and in a small number of games like the playoffs luck is a big factor. Stats can tell you who is better (based on the things they measure, things like injured players aren't accounted for in DVOA. All stats have some blind spots. Knowing where they are helps you use them better). If you know who is better you know who will probably win. Like the guy here did (http://www.bbnflstats.com/) picking 70.8% of this years NFL games correctly.

Football is too complex to be perfectly captured by stats, but DVOA and other stats are useful tools and become more useful every year as they are tweaked to improve their predictive power.

my blog http://shakennbaken.blogspot.com/

by shake n bake on Jan 2, 2008 2:22 PM EST up reply actions

Also about Smith

Of the guys at FO he is the most likely to go away from using stats.

my blog http://shakennbaken.blogspot.com/

by shake n bake on Jan 2, 2008 3:37 PM EST reply actions

Stats

I like stats but obviously they are not the single area that one needs to look at but for anyone to imply stats don't matter or mean anything (some imply this) is silly and usually say this when who/what they like isn't leading in the stats.

I will say that I'm more an individual stats guy than team stats because I know they don't matter that much and can change drastically if a team is up large and playing soft, etc. Take the current year...does anyone truly think the Colts O isn't the 2nd best (or maybe even the best when healthy) in the nfl? Stats say they're lower than this but logic says they are 2nd or possibly 1st.

Oh and judging the playoffs based on stats (or even record) is very silly and people should grasp the simple concept it almost never works for ANY TEAM. I guess though it is easier for people to look at a record or ypg figures and make an opinion on that rather than look deeper and see what is more likely the reality...ie, the Colts may have gone 13-3 but it is clear as day they played at a higher level all season than any other team when you factor in the injuries and such.

Oh and the 06 colts didn't have the worst rush D in regular season history but they were one of the worst, no question. A Bills team from I believe the late 70s (or early 80s) has the honour of being the worst.

by Rob L on Jan 2, 2008 5:53 PM EST reply actions

Misuse of DVOA

It isn't meant to show who will win in the postseason, it is meant to show who played how well during the regular season. I'd compare it to traditional stats (such as YPG) since it measures what has happened.

I think you would agree that the Indy Colts of 2005 were a better regular season team than the 2005 Pittsburgh Steelers. The Steelers won the Super Bowl. Here is the DVOA from that year.

Hogs Haven

by Skin Patrol @ Stampede Blue on Jan 3, 2008 2:27 PM EST reply actions