≡ Menu

Thoughts on Tony Dungy and the Hall of Fame

In the Hall of Very Good Mustaches

In the Hall of Very Good Mustaches

Tony Dungy was selected for enshrinement into the Pro Football Hall of Fame on Saturday. Dungy is the 23rd head coach selected to the Hall of Fame: among that group, he ranks 12th in wins with 139, 9th in winning percentage at .668, and 7th in wins over .500. Those are all impressive numbers, given the sample; the “worst” mark on his resume would be the lone championship, which places him in the bottom six among Hall of Fame coaches (John Madden and Sid Gillman each won one; George Allen, Marv Levy, and Bud Grant won zero titles).

Dungy entered the league in 1996. Excluding Bill Belichick, who is clearly the best coach of this era, where does Dungy rank among the other Super Bowl-winning head coaches? Is he the best choice for the Hall of Fame among this group?

Statistically speaking…. yes. Dungy ranks 5th in wins among this group, but first in winning percentage (in fact, his winning percentage is even higher than Belichick’s!). Perhaps most importantly, he ranks first in wins over .500, which blends raw wins and winning percentage. Coughlin and Shanahan have two rings, but both have combined to win just 52 games more than they have lost; Dungy himself is at +70.

CoachWLTWin%Win ov .500PWPLPwin%Pwov.500
Tony Dungy1396900.668709100.474-1
Mike McCarthy1045510.65349870.5331
Mike Tomlin925200.63940650.5451
Bill Cowher1499010.623591290.5713
Sean Payton875700.60430640.6002
John Harbaugh775100.602261050.6675
Mike Holmgren16111100.5925013110.5422
Pete Carroll936700.58126960.6003
Brian Billick806400.55616530.6252
Mike Shanahan17013800.55232860.5712
Jon Gruden958100.54014540.5561
Tom Coughlin17015000.531201270.6325
Dick Vermeil12010900.52411650.5451
Gary Kubiak736800.5185520.7143

In the playoffs? That’s a different story. Among Super Bowl-winning head coaches since 1996, Dungy is the only one with a losing record. That’s a pretty big negative. On the other hand, 9-10 isn’t appreciably different from McCarthy, Tomlin, Vermeil, Gruden, Holmgren, Shanahan, or Payton.

Dungy was not good in the playoffs, but that’s true of other great coaches. Paul Brown — yes, that Paul Brown — went 9-8 with better teams. In addition to Levy, Grant, and Allen (the three Hall of Fame coaches without championships), guys like Shula and Gillman have pretty uneven playoff records, too.

What I think carried the day for Dungy was — unknowingly, of course — his success in the Dungy Index. You can read more about what that measures at that link, but here were his yearly results:

YearTeamExp. WinsWWin Over Exp
1996TAM6.496-0.49
1997TAM6.73103.27
1998TAM8.618-0.61
1999TAM8.31112.69
2000TAM8.66101.34
2001TAM9.639-0.63
2002IND7.25102.75
2003IND8.52123.48
2004IND9.31122.69
2005IND9.75144.25
2006IND10.34121.66
2007IND8.81134.19
2008IND10.24121.76
Total2TM112.6513926.35

Year after year, his Colts avoided regressing to the mean, much like Belichick’s Patriots. Consistently keeping a team that far above average is obviously a lot easier with the greatest quarterback of all time, but you won’t find many Hall of Fame coaches that didn’t have great quarterbacks. If you want to say that Dungy shouldn’t be in the Hall of Fame because he had Peyton Manning and only won one championship, that’s fine.

But winning multiple championships is obviously not a prerequisite. And among coaches that won only one — or zero — titles, Dungy’s record does stick out as pretty remarkable. Winning over 2/3s of your games is pretty much an automatic pass to the Hall of Fame: every coach with at least 90 wins that has done that is in the Hall. Drop it to 50 games, and the only coach not in there is Blanton Collier, who like Dungy, only won one title. But he coached only one team — the stacked Cleveland Browns — and coached for only 8 years. Given Dungy’s success with multiple teams, he seems like a pretty good Hall of Fame choice.

  • OK, maybe, but I’m still not over his unacceptable approach to the 2007 regular season finale, which dishonored the sport. http://brownsplainly.com/blog/2007/12/hey-tony/

  • sacramento gold miners

    Interesting to note Dungy hired Mike Tomlin for his first NFL job in 2001, and Tomlin may one day join Dungy in the HOF. And Dungy did play QB for the Steelers once in a regular season game. During a 1977 loss at Houston, Dungy was 3-8, 43 yards and two picks, after replacing the injured Terry Bradshaw and Mike Kruczek. As a defensive back, he also intercepted Dan Pastroni during this game.

    Dungy did have only one losing season as a head coach, and was a pioneer in terms of his achievements as an African-American head coach.

  • Dave

    Would be nice to have Belichick’s name and stats in there.

    • Pretty easy to calculate: He’s 223-113, so +110, and at 0.664 winning percentage. 23-10 in the playoffs, so +13 and 0.697.

      • Dave

        Thx. I know we’re not comparing Dungy to Belichick; that we’re comparing Dungy to other SB winning coaches, however, Belichick is one of those coaches.

        • sacramento gold miners

          Had the original Cleveland Browns not moved, and Belichick remained the head coach, I do believe he would have been very successful. However, I think it would have been very unlikely Belichick would have won a Super Bowl with Vinny Testaverde at QB. Belichick’s last playoff game with Cleveland featured a 13-31 fiasco by Vinny, which brought to mind the 1987 Fiesta Bowl disaster.

          • Independent George

            You’re forgetting that he Belichick was the Jets DC when a 35 year-old Testaverde took them to the AFCCG.

  • Richie

    I just never remember thinking “here is one of the best coaches ever” when he was an active coach. Maybe that was my mis-perception. Or, I wonder if his quiet coaching method just didn’t make him stand out to me.

    The string of 12-win seasons is impressive. However, it should be noted, that none of them came in Tampa Bay where he lacked a HOF QB.

    It’s always tough to try to untangle coach performance (or any other position, really) from his QB’s performance.

    I wish Brady or Belichick could get a couple years on the Patriots after the other one leaves, to see if we can separate their accomplishments. But if Belichick is the one to go first, it could just be age that brings Brady’s performance down (if it does).

  • That Paul Brown stat is misleading. First of all, let’s break it down: 9-5 with the Browns and 0-3 with the Bengals. I’m not sure how much I want to beat up a 65-year-old coach who took an expansion team to the playoffs in just their third season. Saying that Brown had a similar record with better teams, I don’t think is correct. He had a far better record with better teams (.643), and a small-sample (n=3) worse record with far lesser teams.

    But more importantly, we in the analytics community have got to stop treating postseason records from the pre-merger era, when almost all playoffs were championship games, the same way we treat postseason records now, when a champion probably got at least one tune-up against the 8th-best team in the tournament. Going 0-1 in 1953, and losing the championship game, is not the same as going 0-1 after losing a divisional game in 2007.

    Along similar lines to the first comment, I don’t think it’s fair to give Dungy anything but credit for what he did in Tampa. The Bucs had zero winning seasons from 1983-95. They went 6-10 in Dungy’s rookie season, and then at or above .500 for the next five years in a row. He made a bad team, with a bad organization, into a good team. His playoff failures with the Colts merit far more criticism.

    • Sure. And I don’t disagree with that.

      OTOH, that 9-5 record is boosted from the AAFC. Here is what I wrote about Brown in my Schottenheimer Index:

      In the NFL, his playoff record was 4-8, and half of those wins came in 1950, the Browns first year in the NFL. Even ignoring his time in Cincinnati (0-3), he went 1-4 in games with a three-point spread and also lost to a far inferior Lions team 17-16 in the 1953 title game. Cleveland started that season 11-0, lost a meaningless final game, and then lost in Detroit. The 1951 Browns were just as good, but also allowed a fourth-quarter comeback to lose the title to the Rams. And despite being better than Detroit again in 1957, Cleveland was throttled 59-14. On the plus side, Cleveland did exact some revenge against Detroit in the middle, winning 56-10 in the 1954 title game.

      I think when you factor in the strength of his teams, his playoff record doesn’t look so good. Not as bad as Dungy’s, sure, but then again, these things are often decided by razor thin margins. You could argue that Dungy’s Colts lost to better teams in ’02, ’03, and ’04, and all of those games came on the road. Yes, the ’05 loss was bad, but Pittsburgh went on to win the title. Those Indianapolis teams lost to the Super Bowl champs in ’03, ’04, and ’05, and then won it in ’06. The SD loss was bad, but Brown lost a game 59-14.

      Dungy’s teams obviously had playoff struggles, but I don’t think that makes him a bad coach. Tom Coughlin’s teams have been very good in the playoffs, but I don’t think that makes him a better coach. Where would you stand on Coughlin/Dungy?

      • Richie

        You didn’t ask me, but I think I would put Dungy slightly ahead of Coughlin. Coughlin may have had more playoff success, but he also missed the playoffs regularly. He went to the playoffs in just 9 of his 20 seasons, while Dungy made 11 of 13 seasons.

        Coughlin may have won 2 Super Bowls, but 2 of his best Giants teams (2005, 2008) lost in the first round. And one of his other best teams (1999 Jaguars) lost the AFC Championship game at home as 7-point favorites.

        I don’t give a whole lot of credit to post-season success and failure, but I think Dungy and Coughlin are similar (including both winning Super Bowls with less than their best teams) in terms of playoff success.

        Dungy was more successful in the regular season. They both essentially started out by coaching expansion franchises.

        • Tom

          I’d put Dungy ahead of Coughlin as well. And as you mention below, it is indeed hard for us as fans to really discern if a team is winning mostly because of the coach, the players…heck, the offensive coordinator? Even so, it’s worth discussing. I came up with a goofy system last year where I weight the playoff games – the SB is worth 8 games, Conference game 4, Division 2, Wild Card or bye 1. Than I calculate a “weighted” W/L%, including regular season games and playoffs (I’m sure this has been done before, it’s like a Bill James-type thing).

          Dungy: 164 weighted wins / 253 weighted games = 0.648
          Coughlin: 208 Ww / 373 Wg = 0.558

          It’s pretty rough, but the intent is to consider both playoff and regular season success.

          In any event, I think he belongs in the Hall for all the reasons every one has mentioned.

          • Using this weighted wins formula, does a coach who loses the Super Bowl get credited for 8 losses? I would’t agree with penalizing a coach for making the big game and subsequently losing, but I don’t know how exactly your calculation works so I want to avoid jumping to any hasty conclusions.

            • JeremyDeShetler

              I believe that is how he is calculating it.

              253 weighted games = 208 regular season games + 45 weighted games in the playoffs (15 in 1 SB appearance, 14 total in 2 Conf Champ appearances, 12 in 4 divisional losses, 4 in 4 wildcard losses)

            • Tom

              Bryan – right, that’s how I’m doing it, and yeah, it’s probably too brutal. Marv Levy is 11-8 in the playoffs for 0.579, his weighted playoff W/L would be 33/74, or 0.446. There’s a bunch of ways to fix this, maybe give half credit for playoff losses, or just change the weights altogether…maybe 4-3-2-1 would be better than 8-4-2-1. I was trying to mimic Chase’s post on Leveraged Playoff performance (too lazy to find the link), where each playoff game is “worth” twice as much as the last. It is my intention that the Super Bowl be weighted pretty heavily though.

              I think it works better for long careers where you can just add up the wins. Belichick has the most playoff wins of any coach, 23 (since 1950), Lombardi has 9, which puts him at #16. If we weight the games though, Lombardi shoots to number two with 76 weighted wins, Belichick is still #1 with 89 weighted wins…Shula is #3 with 73 and Paul Brown is #4 with 68. To me, this feels right – guys get credit for the days when they didn’t have many (or any) playoff games, etc. Of course, doing it this way favors guys who have longer careers, but there’s ways around that as well, etc.

              Probably more than you needed to know, but that’s my line of thinking.

              • I think I would just use the same multipliers you suggested, but only count losses as a single loss. So Levy can be 0-4 in the Super Bowl instead of 0-64 or 0-32 or 0-2, depending on your variable.

                • Tom

                  Bryan – that’s a pretty good idea. So, if you win the SB, you get the 8 wins (and the 8 games is also in the denominator, as it were). If you lose, it counts as just one game…Marv’s weighted playoff W/L would jump to 0.717 (33/46). I’m OK with that…gives the guy credit for four AFC Championships, which is pretty good.

                  • I think the .717 weighted score comes much closer to matching what we think about when we think about Levy.

      • I rank Dungy ahead of Coughlin. I think Dungy seemed out-matched in big games, and he was routinely out-coached in the playoffs, but I think the points in his favor outweigh the negatives. I don’t want to do a massive copy-and-paste, but I wrote in depth about Dungy here if you’re interested in the details of my reasoning.

        I’m not sure why Brown’s 5-0 record in the AAFC doesn’t count. Those were postseason games in a major league. Surely beating the 10-3-1 Yankees in the 1946 Championship Game is more important than beating the 9-7 Chiefs in the 2006 wild card. Why count one but not the other?

        I get the impression that you’re playing devil’s advocate a little bit, starting with Brown~Dungy, and working backwards looking for evidence to support that point, rather than beginning from a neutral point and going where the evidence takes you. You discount the AAFC years, count wild card wins being just as impressive as championship wins, and count wild card losses being no worse than championship losses. If you do all that, sure Brown looks a little like Dungy. But that’s not an accurate representation of what they both did.

    • sacramento gold miners

      Not picking Bill Walsh to succeed him as head coach of the Bengals was a curious move. And in the Bill Walsh documentary on the NFL Network, it was claimed Brown talked negatively about Walsh when the latter interviewed for other NFL jobs.

      • True, Bill Walsh greatly resented Paul Brown for not supporting his head coaching ambitions. Brown was the greatest coach in pro football history, and I believe this is so obvious that anyone who feels otherwise is clearly underinformed, but when I wrote about NFL coaching trees, I put Walsh in the Al Davis branch, not Paul Brown.

  • For those curious, Manning was the starting QB in 125 of Dungy’s 227 games (55%), Trent Dilfer for 60 (26.4%), Shaun King for 24 (10.6%), Brad Johnson for 17 (7.5%), and Eric Zeier for 1 (0.5%).

    His winning percentage was 0.736 with Manning, 0.533 with Dilfer, 0.625 with King, 0.529 with Johnson, and 0.000 with Zeier.

    • sacramento gold miners

      That’s a revealing stat, some fans would think the percentage of Manning games with Dungy would be greater. Being successful with two different NFL teams is something Marv Levy and Hank Stram weren’t, so that adds value to Dungy’s case. A solid hall of famer, and Dungy did have success as a coordinator before that. His 1984 Steelers defense held the great 49ers team to their second lowest point output that season as Pittsburgh upset San Francisco at Candlestick Park.

  • Adam

    Dungy’s stint in Tampa Bay was HOF-caliber, but not long enough to justify induction on its own. I don’t think Dungy deserves much credit for his time in Indy; the team’s strength (the offense) was run by Peyton Manning, while the defense (Dungy’s specialty) was generally mediocre, with 2007 being the one exception.

    Tony Dungy, HOF = Jerome Bettis, HOF

    • WR

      So you weren’t impressed by Indy’s defense when they finished 7th in fewest points allowed in 2002, 2nd in 2005, or 7th in 2008?

      Also, if Manning was running the offense, why did the Colts bother to employ an offensive coordinator?

      • Adam

        The Indy defense faced the fewest drives in the league during Dungy’s tenure, so their points per drive allowed was significantly worse than their points per game. Also forced relatively few turnovers, which IMO is perhaps the most vital task of a defense.

        Tom Moore was a figurehead OC; his job was to give Peyton “concepts”, but Peyton was in charge.

        • WR

          Can you provide some links and quotes to substantiate the claim that Moore was just a figurehead? I found an article that largely dispels that myth

          http://espn.go.com/blog/green-bay-packers/post/_/id/21092/play-calling-for-aaron-rodgers-peyton-manning-doesnt-and-neither-will-he

          And if Manning really was both the QB and the assistant coach of those teams, doesn’t he bear even more responsibility for the playoff failures those teams had under Dungy? The 41-0 loss to the Jets, the afc champ in which Manning threw 4 costly picks, the 20-3 loss from 2004, and the loss to the Steelers

          • Adam

            I heard Cris Collinsworth talk about this on SNF a few years ago, plus it seems to be the general consensus that Manning ran the offense. But let’s not turn this into another argument about Manning, because that’s not the point I’m trying to make. Even if, hypothetically, Tom Moore was the mastermind behind the offense and Manning was just along for the ride, that doesn’t change the amount of influence Dungy had on the offense – which I believe to be very little. Dungy’s offenses in Tampa were generally poor and never above average; he’s a defensive coordinator masquerading as a head coach. Given that his speciality is defense, I hold him accountable for not turning the Colts’ D into an elite level unit. The Colts also had abysmal special teams during the Dungy regime, which he deserves some of the blame for.

            My point is, Dungy seems rather useless in the realms of offense and special teams; therefore, he better be absolutely nailing it on defense. He did so in Tampa but not in Indy. I also think Dungy was too conservative with his in-game strategy, and his obsession with resting starters was somewhat detrimental to his team.

            IMO, Dungy’s closest comps are guys like Cowher, Fox, and Schottenhemier. These are solidly above average coaches who could stabilize a franchise, but I don’t see anything particularly special about them. Hall of Very Good, but not Hall of Fame.

            • Richie

              I agree with your summary, however I also expect Cowher to be inducted into the HOF within the next decade.

            • That seems like a very limited idea of what a coach does. To call Dungy useless in the realms of offense and special teams is probably a big overstatement of what you mean, which is that he’s not designing offensive plays or special teams plays. But a head coach is responsible for the whole team, which means all the coordinators, assistant coaches, and players. It’s hard to imagine that Dungy’s fingerprints weren’t on the whole team, even if he’s not drawing up Xs and Os.

              • Adam

                Yeah I didn’t mean that to come across so harshly. I just think his impact on offense and special teams was relatively minimal compared to the defense. He could be a great leader of men who gets the most out of his players, but I don’t know how we prove that either way.

    • sacramento gold miners

      I would go along with the Dungy=Bettis comparison if Dungy ranked among the highest ever in terms of victories, but he gets in for other reasons. Since you brought up Bettis, I did a little research which proves how significant his age 32 productivity was in 2004. Since 1990, he’s the only RB with six 100 yard rushing games at that age or older. Better than Emmitt Smith, Marcus Allen, and other standout backs. Compilers aren’t suppose to play this well in their next to last seasons. Lots of HOF backs would have loved that kind of production at that stage of their respective careers.