PDA

View Full Version : Measuring a team's effectiveness against the run


QuikSand
02-10-2008, 06:41 PM
I have been fiddling a little, and have been looking for a simple way to measure how good a team is at stopping the run. What I find challenging about doing this is that "playing the run" is something that a team tends to do less and less of once its record gets good -- when you're winning games, you tend to get leads, tend to face teams who need to pass, and sit back and expect the pass. The result can be a deceptive stat -- you might be giving up a fair number of 5 yard gains on 3rd and 9... which would drive up a stat like yards per carry allowed... but that's still a good result for your team. Anyway -- I have often noticed that it's tough for a winning team to maintain very good numbers in the ypc-allowed stat, and I assume it's sizable for this reason, they are just expecting the pass more often.

Then, I was flipping through the team stats section, and it hit me. There's a backwards way of getting at something very much like what I really want here. Recall an earlier discussion that revealed the nature of the "key run block" stat -- it's essentially a measure of whether a designed run play earned a mark of "success" by some metric (my impression is that it's at lest 5 yards on first down and any distance, on second down it's some share of the remaining yardage toward a first down, and on third or fourth down only a first down or TD qualifies as a success). So, since in FOF team stats are all inverted for a team and its opponents, then looking under "Opposing Linemen" should let you know how your team did on defense.

How to look up this stat:

-Go to the Almanac window
-Choose Team Statistics
-Select Opposing Linemen
-Click the top of the column labeled BPct

Now, you have a sorted list of how often each team's defense gave up a successful running play, which isn't an altogether bad way to try to set aside some of these situational differences, and put rushing defense into context.

It's not perfect -- I don't know if "success" is perfectly defined, and certain types of plays likely have a better average success rate that do others. It also diminishes the effect of single big plays... giving up one 70 yard TD run can really change a game, but in this stat it won't hurt you more than giving up one 4-yard first down plunge on 2nd and 4. But this isn't a bad stat, I don't think... it might be the single best consolidated stat in-game to measure a team's overall rushing defensive effectiveness.

Tuck it away.

JetsIn06
02-10-2008, 07:14 PM
Wow, very interesting stuff to think about. I've always thought that YPC was a faulty measure for some teams that are consistently in the lead.

Here's an interesting take from the WOOF. I rank 2nd in opponents YPC, giving up only 3.5 YPC. However when I go to the stat you pointed out, I'm right in the middle of the pack.

The interesting thing here is that I'm not a bad team that is getting consistently run against to run out the clock, so I think I'm legitimately stopping the run pretty well. So, how should I interpret the data?

QuikSand
02-10-2008, 07:38 PM
I've been thinking about this, too. Maybe an amalgamation of the two stats (OppYPC and OppRB%) would be a better combined stat.

Of course, in one season, if a pretty good run defense just happens to give up, say, three or four long TD runs on bad dice rolls or whatever... then your pretty-good stats can rise by about 0.4 or 0.5 in ypc, and that can easily be the difference between being in the top versus the third quartile.

JetsIn06
02-10-2008, 07:49 PM
One thing that I think I can interpret from my stats is that the opposing offense is getting into a lot of 3rd and short situations. Yea, I'll give up about 1-3 yards on first down. Then I'll give up a pass of 6-7 yards. Now it's 3rd and 1 or 2. They should be seeing more "successful" run blocks because they're converting. My YPC stays low because they're still only picking up 2 or 3 yards.

MIJB#19
02-11-2008, 08:29 AM
Great find, QuikSand.

JetsIn06, I think you just brought up one of the things that's missing in the huge set of stats called Team Statistics, whereas individual players are tracked for their third down running, passing and receiving, teams are only tracked for 3rd down offense and defense, but not the seperate 3rd down run and pass offense and defense. Without those seperated, it'll be really hard to find a way to factor in the 3rd down 'success' of your run defense.

QuikSand
02-11-2008, 12:25 PM
One thing that I think I can interpret from my stats is that the opposing offense is getting into a lot of 3rd and short situations. Yea, I'll give up about 1-3 yards on first down. Then I'll give up a pass of 6-7 yards. Now it's 3rd and 1 or 2. They should be seeing more "successful" run blocks because they're converting. My YPC stays low because they're still only picking up 2 or 3 yards.

Well, *if* that really is true, then I suspect that would be a source of noise in the success rate. I have trouble imagining any reason why a given team would have a systematically higher or lower share of 3rd-and-short situations, though. It doesn't seem to me like there is a lot of magic in the totals of yielding 8 or 9 yards in two plays, as opposed to 7 or 10.