The Results are In: A Ridiculously In-Depth Analysis of the Mike Tomlin Era

Photo via Steelers.com

This post was not only inspired but requested by Steeler Fever after I published the following in a post last week:

  • Mike Tomlin overall regular season record prior to 2017: 103 – 57
  • Record against opponents who made the playoffs: 27 – 29
  • Record against non-playoff-achieving opponents: 77 – 27
  • Record against teams with a winning record: 46 – 36
  • Record against teams with a losing record: 57 – 21
  • Record against teams who finished with a better record that the Steelers: 17 – 18
  • Record against teams who finished with the same record as the Steelers: 9 – 11
  • Record against teams who finished with a worse record than the Steelers: 78 – 27
  • Record against teams who finished with a 4-12 or worse record: 24 – 6*

Fever asked me to make it a separate post so he could use it to vanquish the nay-sayers, or words to that effect. So I agreed to do so, but with considerable amplification, since I had put the above together rather hastily, and since other ways of looking at the data had occurred to me.

I entered everything into a spread sheet, which I should have done in the first place, made a separate category for 8-8 teams, which I also should have done in the first place, looked at the numbers from various different angles, and saw some really interesting stuff, at least if you like numbers. So here goes:

The first thing which strikes me is that not only has Mike Tomlin never had a losing season, he has never had a team with a negative scoring differential. You might think “Well, duh!”, but it isn’t as obvious as that. A team who manages to scratch out enough wins to get to 8-8 but loses a number of blowouts could easily end up with a negative scoring differential, even a large one. Not too surprisingly, the closest the Tomlin-era Steelers ever got was in their two 8-8 seasons, 2012 and 2013, when their scoring differentials were +8 and +9 respectively. The next closest they ever got was in 2009, where the differential was +44. In the other seven seasons they never had less than a +50 differential, and in four of the years the differential was over +100.

Before I continue with the other information, here are some things I think are worthwhile to bear in mind:

1. I proceeded under the assumption that the only way to know what teams the Steelers “should” beat is in retrospect. We had a classic example of that last season. Steeler Nation was quite exercised when the Black and Gold traveled to Miami to play an 0-5 team and lost. With the benefit of hindsight we now know that 0-5 team didn’t lose another game until they lost to the Steelers at Heinz Field in the playoffs. So were they a bad team? They had been a bad team up until that point, and in fact the head coach spent the preceding week shaking things up in the locker room, which generally has a galvanizing effect upon those who remain. There was plenty of talent on the team—they just hadn’t been getting it done. (And see point No. 3.)

2. Not every Steelers team is going to win a Lombardi, nor is this a reasonable expectation. I think many in Steeler Nation are realistic about what a given team in a given year and given circumstances can do, and the best thing would be to calculate all of those factors and then decide if said given team is performing up to or below reasonable expectations. This does not mean always beating teams they “should.” That’s why there are “upsets” every week. If a team has too many “upsets” it is reasonable to assume that, in fact, they really weren’t a very good team to begin with.

3. The previous statement may seem like a circular argument. But it is easy, especially for those who are deeply embroiled in the fantasy football culture, to forget that a team is much more like an organism than it is a random collection of people. How they ended up on the same team may be, to some degree, random, but once they are the team becomes a collection of overlapping interpersonal relationships. On a good team they have a single primary goal—to win a Lombardi—and well-aligned individual goals, such as to become better, stay healthy, work well with their teammates, and so on. And add to this all of the various relationships between what you might call labor and management, and the relationships among the coaching staff and the front office and the ownership, etc., it’s all very complicated.

Personally I would put a lot of what one might consider underperforming (on a season-wide level) on this cauldron of relationships. This is, of course, in addition to factors which may or may not be obvious to the general public—players playing more hurt than perhaps they should (like Mike Mitchell in his first PIT season,) players and/or coaches dealing with heartache or ill/dying family members or other situations where real life makes football seem less important, or you name it. I can think of dozens of scenarios we might never know about which effects individual players or whole locker rooms. And those are in addition to some of the ones we can see, like suspensions or holdouts or players “quitting” on the team or what have you.

5. All of which is why you can’t calculate X amount of talent, Y amount of schedule difficulty, Z amount of injuries (or lack thereof), combine it with an X% effective coaching staff, and have the results pop out. If it were possible, Vegas would never lose. And even in retrospect you can never be entirely certain about these things.

And finally:

6. Stats in football are difficult. In baseball you have 162 games per year in the regular season alone. It takes over 10 football seasons to reach the stats you get in a single season of baseball. As this is Mike Tomlin’s 11th season as head coach with the Steelers, we are at the exciting point where we have about the same number of data points as a baseball team would have had after one year. I do think it is enough to see trends, though. But little trends (like supposedly “playing down” to good teams) definitely have to be viewed in terms of more global trends, such as a defensive core growing old together

The popular concept of the Steelers is that they take bad teams very lightly, or “play down” to them. In the list above I just looked at the record against 4-12 or worse teams, and as you can see there are very few losses

So does this mean you can’t judge a coach by his record? I’m sure that none other than Mike Tomlin would scoff at the very idea. As he would say, “You are who your record says you are.” And while a lot of things are not under his control, in a general way, that’s true. So let’s see who he really is according to his record:

Win/loss total: 103-57. Win %: .644 (Note: I am only considering the regular season. The playoffs are a much smaller sample size with a great deal more variability.)

Here is a comparison to all the other active head coaches who were head coaches during most of the past decade. (Note that this means the coaches had some reasonable measure of success to have lasted that long, whether with one or more teams:)

In case it makes anybody feel better, Bill Belichick’s winning percentage prior to 2001 was .427…

Now on the the discussion about wins against teams with winning/losing records. There are two facets of this—the records of the two teams playing at the time of the game and the records of the teams at the end of the season. And although the “time of game” records become somewhat moot towards the end of the season, and don’t mean very much right near the beginning, it is still interesting.

So I noted for every single game (yes, all 160 of them) whether the Steelers had a better, equal, or worse record than the team they were playing, both at the time of the game and at the end of the season. This proved very useful for determining in a somewhat more objective way what could be considered to be game the Steelers “should” win. I could give you those raw numbers, but I suspect it is more interesting to go straight to the calculations.

You may not agree with these designations, but here is what I decided on. I didn’t look at how it would fit the data before I determined my categories, in case you are wondering, because that would be unscientific. Don’t forget—any terms in quotes are now technical terms, for the purpose of this article:

A “Should Win” game is a game against a team whose record is worse than that of the Steelers both before the game begins and at the end of the season.

A “Bad Team” is a team who ended the season with a 4-12 or worse record.

These two terms are used to determine what is a bad loss. Heck, let’s break down the “bad loss” category even further. Here’s the explanations:

A “Dismal Loss” is a “Should Win” loss at home when the opponent ends the season as a “Bad Team.” (In my opinion, no road loss, however ugly, is in the same category as a home loss, particularly given Ben Roethlisberger’s home/road splits.)

A “Bad Loss” is a “Should Win” loss on the road to a “Bad Team,” OR a loss at home to a team who ends the season with a losing record (unless said team makes the playoffs.)

But let’s not leave it at this. Shouldn’t we also consider whether some wins are better than others? We all know a win at Gillette Stadium is, as the ads say, priceless. But let’s attempt to put a price on a couple of other sorts of wins:

A “Good Win” is a home win against a team who has a better record than the Steelers, both at the time of the game and the end of the season.

A “Great Win” is the same, but on the road.

And finally I offer a couple of categories which may never make the prime time, but seem worth looking at. I give you:

A “Good Loss” is a home loss by a field goal or less (these are frequently in overtime) where the opponent has a 12-4 or better record, or a loss by less than a touchdown on the road to a team with a better record, both at the time of the game and the end of the season.

A “Bad Win” is a win by a field goal or less, at home, to a “Bad Team.”

Yes, there is no such thing as a good loss or a bad win, but sometimes the former can make you feel better, and sometimes the latter are straws to show which way the wind is blowing.

I almost decided to have a separate category for Ravens games. Win or lose, they always come with a bit of an asterisk in my mind. But they all count the same on the scoreboard. So this will do it. And now to definitively and scientifically decide whether the Steelers under Mike Tomlin “play down” to lesser opponents, and if so whether this is at all balanced out by “playing up” to better ones. Because I notice nobody ever complains about that, except perhaps the fan base of the team they beat…

“Dismal Losses”— one. Yes, you read that correctly. During the entire Mike Tomlin era there has only been one loss to qualify—the 2014 home loss to the 2-14 Tampa Bay Buccaneers, quarterbacked by the self-same Mike Glennon who beat the Steelers in Chicago last Sunday. None of the other losses during his tenure, bad as they may have seemed at the time, qualified. (Note in the list at the beginning of the article that Tomlin has only six losses to team who ended the season at 4-12 or worse.)

“Bad Losses”—eight. Note I broadened the definition of a “bad loss” at home from just losses to 4-12 or worse teams. This gives a number which is probably more in line with fan perceptions. Although this averages out to under one per year, in fact they clustered in 2009, 2012, and 2014, with one other outlier in 2015.

If we add “Dismal” and “Bad” losses together, this is 9 out of 160 total games, for a Total Bad Losses rate of .056.

Now for the good news. The Tomlin-era Steelers have the following:

“Great Wins”—five. Three of those came in 2012, when their record hit a nadir, not too surprisingly.

“Good Wins”—seven. Three of those came in 2009.

And in case you’re wondering, I’m not double-counting. A win can only be either a “good win” or a “great win.” Same for the losses…

And just for kicks, the Steelers have, according to my metric and calculations, four “Bad Wins” and two “Good Losses.” As Craig Wolfley says, sometimes it’s better to be lucky than good.

Finally, let’s look at these altogether:

“Dismal Losses”—1 in 160 games.

“Bad Losses”—8 in 160 games.

“Good Wins”—7 in 160 games.

“Great Wins:—5 in 160 games.

In other words, under Mike Tomlin the Steelers have won 25% more games in which they were overmatched than they have lost to teams they “should have beat.” At least if we are realistic about who they should or shouldn’t beat. Because in the end you are who your record says you are.

I would love to account for injuries, particularly to critical players like, say, Ben Roethlisberger, but then you get into the whole “a win against Case Keenum is somehow less worthy that one against Sam Bradford” argument. (And Case Keenum looked pretty darn good in the Vikings’ 34-17 win against Tampa Bay on Sunday. He was 25 of 33, 3 TDs…) Although some players have an outsized degree of importance, in the end you win or lose as a team. This was never better illustrated than in the Steelers’ 3-1 start to the 2010 season during Ben Roethlisberger’s suspension.

And were I to make allowances, such as in the dreadful (but not “dismal”) loss to the 49ers back in 2011, I would have to make the same calculation for the opposing teams. All 160 games’ worth. And as a wise woman once said, “Ain’t nobody got time for that.”

*I miscounted in the original article, and it was given as 24-5.

5 comments

  • This a terrific, enlightening piece. Always nice when you bump perception up against data.

    Like

    • Thanks, Ivan. There are many other possible ways to look at the data, because I set the criteria and others might not agree with them, but I tried to do so in a not-too-homeristic fashion, and be reasonable (tough as a fan) about what one should expect. Numbers will never tell you much about the intangibles, but even for those numbers tell you something…

      Like

  • Those numbers say a lot about the Tomlin era. There is a question though as to how the other coaches you show above fair in the same metric. If you were to send me your spreadsheet, i would be happy to crunch the numbers for you.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s