PDA

View Full Version : Stats can mislead


keyser
09-17-2007, 10:14 AM
This is a kind of random post about something I found interesting... Generally, I like stats, but when looking over them after this last game, I was reminded just how misleading they can be. Here are some of stats for the Carolina game:

1st downs: Houston - 19, Carolina - 17
3rd down efficiency: Houston - 6/14, Carolina - 7/13
Total Yards: Houston - 346, Carolina - 344
Penalties: Houston - 5/40, Carolina - 5/35

True, other stats like turnovers (3-1) give a better picture, but from the above, you'd think this was a lot different game than it was...

Kaiser Toro
09-17-2007, 10:18 AM
This is a kind of random post about something I found interesting... Generally, I like stats, but when looking over them after this last game, I was reminded just how misleading they can be. Here are some of stats for the Carolina game:

1st downs: Houston - 19, Carolina - 17
3rd down efficiency: Houston - 6/14, Carolina - 7/13
Total Yards: Houston - 346, Carolina - 344
Penalties: Houston - 5/40, Carolina - 5/35

True, other stats like turnovers (3-1) give a better picture, but from the above, you'd think this was a lot different game than it was...

That is why we watch the game. I believe there was a 150 yard delta between each team at the beginning of the 4th quarter.

You did not have to look at the scoreboard, because the lack of fans in the stands midway through the 3rd quarter said it all. :texflag:

Twitch-Houston
09-17-2007, 10:21 AM
Yea the stats make the game look closer than it actually was. The Texans dominated the game. I think that long TD where Steve Smith broke tackles helped their (especially Delhomme's passing) stats.

gtexan02
09-17-2007, 10:22 AM
That 70+ yard play to Smith really helped pad delhommes stats.

In garbage time.

YKW's influence must be strong, indeed. "Delhomme, go for the stats only when the game is totally out of reach! They'll play a soft zone and you can look great, while still being a loser!"

BeerTastesLikeVictory
09-17-2007, 10:36 AM
I think we all can agree stats are misleading. Just remember a good portion of the year in 06, D. Carr was sitting pretty with his completion percentage. If you were to take the stats for the first 3 quarters it might tell a different story.

Texans_Chick
09-17-2007, 11:12 AM
This is a kind of random post about something I found interesting... Generally, I like stats, but when looking over them after this last game, I was reminded just how misleading they can be. Here are some of stats for the Carolina game:

1st downs: Houston - 19, Carolina - 17
3rd down efficiency: Houston - 6/14, Carolina - 7/13
Total Yards: Houston - 346, Carolina - 344
Penalties: Houston - 5/40, Carolina - 5/35

True, other stats like turnovers (3-1) give a better picture, but from the above, you'd think this was a lot different game than it was...

Check out the time of possession. Huge difference in that stat. Two games in a row with time chewing drives is a good way to go into the Colts game.

Double Barrel
09-17-2007, 11:15 AM
The only stat that really matters is scoreboard! :texflag:

gtexan02
09-17-2007, 11:23 AM
Check out the time of possession. Huge difference in that stat. Two games in a row with time chewing drives is a good way to go into the Colts game.

That was the huge stat difference against the Colts last year as well.

Hervoyel
09-17-2007, 11:25 AM
When you get up like that you inevitably pull back a little and try to eat up the clock. That always gives the other team drives in garbage time that make things look closer than they were. We should know that considering we got most of last seasons offensive production in garbage time. As much as I respect Smith and the game he has I have to say that even his long TD with the incredible effort he gave was a garbage time stat. He gets points for not giving up but the game was still (mostly) out of reach when he did it.

Jake Delhomme is used to producing when the game is still in question. He didn't do so well once we pulled away from them. The Panthers should have put their "garbage time specialist" in the game and he would have lit it up.

Texans Horror
09-17-2007, 11:41 AM
I think it was infantrycak that had a website that charted "big plays" - turnovers, special teams, that sort of thing. It was a great link, and I think the theory would hold true here. For the second week in a row, the Texans won the big plays.

infantrycak
09-17-2007, 11:59 AM
It is on Football Outsiders. It will be updated in a day or two hopefully.

The Pencil Neck
09-17-2007, 12:34 PM
Yeah, I'm waiting to hit Football Outsiders and Cold Hard Facts sites after they've had a chance to update their stats.

mancunian
09-17-2007, 12:42 PM
The only stat that really matters is scoreboard! :texflag:
and the number of big W's you have.

jlam
09-17-2007, 02:18 PM
That 70+ yard play to Smith really helped pad delhommes stats.

In garbage time.



That actually worked out really well for me. Pad Smith's stats (he's on my fantasy team) while the Texans took out a division opponent.

Two small caveats: We lost to a divisional opponent and I still lost in my Fantasy league. Don't you hate it when you score more points than everyone else in your entire league except for the guy you happen to be playing. Pisser. :(

Hooston Texan
09-17-2007, 03:55 PM
Some games, you have to look at the drive charts rather than the box score to see how the game truly played. Yesterday was a prime example.

While the game was truly up for grabs (until we went ahead 31-14), this is how they did:

Texans: 5 offensive possessions (not counting the kneel at the end of the half): 39 plays, 289 yards, 31 points, 1 turnover.

Panthers: 6 possessions, 26 plays, 111 yards, 14 points, 1 turnover

That means that the rest of the way, the teams did as follows:

Texans: 5 possessions, 22 plays, 58 yards, 3 points

Panthers: 6 possessions, 35 plays, 247 yards, 7 points, 2 turnovers

In fact, 165 of the Panthers' yards came on the last two drives. For the other 52 minutes, they had less than 200 yards in offense. That's getting it done.