Oct 14, 2013; San Diego, CA, USA; Indianapolis Colts full back Stanley Havili (39) on the sidelines during the second half against the San Diego Chargers at Qualcomm Stadium. The Chargers won 19-9. Mandatory Credit: Christopher Hanewinckel-USA TODAY Sports

Stanley Havili’s effect on the running game

Stanley Havili was a lightning rod for criticism in his first year with the Colts, mostly because playing him meant keeping T.Y. Hilton – or, later in the year, someone like LaVon Brazill or Da’Rick Rogers – on the bench. It didn’t help when a pass bounced off Havili’s hands in the playoff game against the Patriots, leading to an interception (something similar happened in the preseason).

Then came the news last week that erstwhile linebacker Mario Harvey was moving to fullback. Havili made a few plays last year, particularly as a receiver, but as inconsistent as he was, it certainly seems plausible that someone like Harvey could push him out of his roster spot.

But how fair is that perception? How useful was Havili? In particular, what effect did his lead blocking, presumably a fullback’s most important trait in Pep Hamilton’s old school offense, have on the Colts’ running game? These are the questions that keep me up at night in the NFL offseason.

So I flung myself into Game Rewind and busted out a couple new features of my block charting system. I marked every run on which Havili served as a lead blocker, meaning he started in the backfield, moved forward and blocked someone on the play-side (or what appeared to be the designed play-side in the event of a cutback; it’s usually pretty easy to tell). I didn’t include plays on which he went to the edge of the formation and kicked his man out, since that’s a different sort of block.

For a comparative sample of conventional running plays, I took every Colts run, threw out kneel-downs and penalties and such, and subtracted weird situations like end-arounds and Tashard Choice garbage time carries. Here are the numbers I came up with:

With Havili lead blocks
Without Havili lead blocks

The touchdowns are slightly more frequent in the Havili block column, but the averages scream that the Colts were more productive without him. Time for some gleeful Havili-bashing, right?

Well, not exactly. As we all know, all yards are not created equal. A 2-yard Andrew Luck plunge on 3rd-and-1 is far more valuable than a 13-yard Ahmad Bradshaw run on 3rd-and-31 (or, say, a 1-yard Havili run on 3rd-and-10 – not that Hamilton would ever do that).

Success percentage is one alternative to yards per carry and often provides a better picture of effectiveness, since it values consistently putting the offense in good positions and devalues big plays. Different sources use different formulas to calculate success, but I’m using a simple and widely accepted one: to be successful, a play must gain 40% of the needed yards on first down, 60% on second down and 100% on third or fourth down.

Using the 40/60/100 metric, the Colts were successful on 162 of the 358 runs in my total sample, or 45%. According to Football Outsiders’ 2013 rankings (which use a slightly different formula), that would have come in between the 31st- and 32nd-ranked players last year. It wasn’t a banner year for the Colts’ running game, but you already knew that.

Here’s where it gets interesting. On plays with Havili lead blocks, the Colts succeeded on 51 of 105 runs (49%). Without Havili lead blocks, they went 111-for-253 (44%). On FO’s list, that’s equivalent to the difference between the 17th-ranked player and the 32nd-ranked player. Plays with Havili lead blocks tended not to be exciting and usually resulted in 3- to 4-yard gains. But whereas runs without his lead blocks often turned into losses, runs with them usually yielded a positive result.

Given that removing a fullback generally means spreading the field with another receiver, you might expect the drop in success to be balanced by an increase in big plays. The Colts did manage longer runs a bit more frequently when Havili wasn’t lead blocking, but the differences were negligible:

With Havili lead blocks
Without Havili lead blocks
Runs of 5+ yards3279111
5+-yard run %30%31%31%
Runs of 10+ yards82533
10+-yard run %8%10%9%

You might also think Havili would play more on 3rd-and-short, inflating his success percentage with a bunch of short gains. But the Colts liked to use guys like Harvey and Ricardo Mathews at fullback in such situations. Havili had only five third-down lead blocks in my sample; the Colts succeeded on four of them, averaging 2.8 yards. Without his lead blocks, they succeeded on only 13 of 32 (41%), averaging 3.2 yards.

Havili’s value showed most on first down:

With Havili lead blocks
Without Havili lead blocks
Successful runs286088
Successful run %47%42%43%
Runs of 5+ yards194160
5+-yard run %32%28%29%
Runs of 10+ yards5712
10+-yard run %8%5%6%

That’s right: on first down, not only were the Colts a more consistently successful running team with Havili as a lead blocker, they produced a higher percentage of runs of both 5+ and 10+ yards, with an almost identical average gain.

Interestingly, second-down runs went better when Havili wasn’t lead blocking:

With Havili lead blocks
Without Havili lead blocks
Successful runs183553
Successful run %46%48%47%
Runs of 5+ yards122739
5+-yard run %31%37%35%
Runs of 10+ yards31316
10+-yard run %8%18%14%

The success percentages were close, but the average and big plays overwhelmingly favored the runs without Havili lead blocks. It’s a relatively small sample, and four of the Colts’ seven runs of 20+ yards in this group of plays came on second down, which skews the average a bit. It might also be an effect of Hamilton’s play-calling tendencies, and in any case the differences weren’t significant enough to balance out Havili’s advantages on first downs.

Havili was particularly useful early in the season, before nagging injuries and a decreasing role limited his value: after a dud in the opener, the Colts were successful on running plays with him as the lead blocker at least 56% of the time in each of his next five games. He only had one game over 50% the rest of the season, but when he was on, the Colts were a markedly better running team.

I’m not arguing that Havili should play all the time. It’s a truism that passing is more efficient than running in the modern NFL, and the Colts’ offense didn’t recover from Reggie Wayne’s injury last year until they started going shotgun/three-wide and letting Luck work the field. You could also certainly argue that a slight drop in running efficiency is well worth it for the more dynamic possibilities of a three-wide or two-tight end set. But if they’re going to keep pushing their running game as they did last year (notwithstanding Hamilton’s recent promising words about making the Colts a score-first team rather than a run-first team), Havili does have value as a lead blocker.