Friday, January 10, 2014

Expanding the Playoffs Part II: Data Analysis

In Part I of my series on examining the NFL's proposed expansion of the playoffs from 12 to 14 teams, I took a look at how history may have been different if the NFL had instituted this policy in 2002 when the league expanded to its current playoff format of 4 division champions and 2 wild card teams per conference. Prior to 2002, the each conference had 3 division champions and 3 wild card teams that reached the postseason. With the proposed expansion of the postseason, there would be 7 teams per conference and only the top seeds would get a bye. For Part II of the series, I look at some of the data for the teams that finished in 7th place in their conference standings over the last 12 years.

In any given season, it is debatable about which teams "deserve" to make the postseason. One of the biggest misconceptions is the perceived "dropoff" from the playoff teams to the teams that did not make the postseason. On average, the 6th seeded team has had 9.75 wins per season while the 7th place team has averaged 9.08. As the second chart above shows, 7 of the 24 teams that finished in 7th place had double-digit wins. Each of these instances occurred in separate seasons meaning that in 7 of the last 12 seasons (more than half of the time) a team with double-digit wins has actually missed the postseason. The first chart shows an interesting trend from a parity perspective: in the first half of the chart (2002-2008) there were 4 AFC teams and no NFC teams that finished in 7th place with double-digit wins while in the second half (2009-present) there were 3 10-win NFC teams and no AFC teams that finished 7th. Additionally, in the first half 3 of the 7th place teams in the NFC were 8-8 (none in the AFC) while in the second half there were two 7th place AFC teams (both the Steelers) at 8-8 and just one NFC team. Clearly, as we have seen on the field, the balance of power in the league has shifted from the AFC in the early-to-mid 00's to the NFC in the early 2010s.

When comparing the number of wins by 6th and 7th place teams, one has to expect some lowering of records (these are teams that finished lower in the standings afterall). Since 2002, only two teams with 8-8 records have finished 6th (though we should also count the 8-8 Rams of 2004 that were the #5 seed at 8-8) while 6 of the 24 8th place teams had .500 records. While some would say this is increasing the number of mediocre teams in the playoffs, we must also remember that 7 of the 7th place teams finished with 10+ wins. In my opinion, the 6 8-8 teams that would have made the playoffs are balanced out by the 7 double-digit win teams that would have reached the postseason, so it is hard to say that the playoff pool is being definitively muddled when that many quality teams with 10+ wins are also getting in.

Interestingly, when taking the rest of the playoff field into consideration, the profile of 7th place teams actually compares closest with #4 seeds. On average, #4 seeds (the worst division champion) have had 9.65 wins, slightly less than #6 seeds. On 10 occasions, the 7th place team had a record better than or equal to the record of the #4 seed in their conference. In 9 other instances the 7th place team finished only a game behind the #4 seed in the overall standings.

Logistically (and regardless of record) it is easy for football fans to break teams into "tiers" based on whether or not they made the playoffs. However, based on the first chart above we see that only 4 of the 7th place teams finished more than a game behind the 6th place team in the standings. In fact, almost half (11 of 24) of the 7th place teams finished tied with the 6th place team but missed the playoffs due to tiebreakers. A majority of those 11 that missed due to tiebreakers (7) finished in 7th place alone and did not require additional tiebreakers to award them the 7th spot in the standings. On the whole (as the second chart shows) a majority of 7th place teams (14) finished in 7th alone and were not tied with the team that finished 8th. This would lend credence to the notion that the "tier" of teams that is at the top of the league is actually after the 7th place team, not after the 6th place team.

One of the most interesting items I found when combing through the data was how the 7th place teams finished the season. 20 of the 24 teams finished the final quarter with at least 2 wins. Additionally, as many teams (10) finished with winning records in the final quarter (going either 3-1 or 4-0) as posted .500 records over the final four games. The success of 7th place teams over the final quarter of the season is borne out when we look at their streaks at the end of the season. A vast majority (18 of 24) ended the season with a win in Week 17 with 10 teams having won at least their final two games. The next question which arises is - how did these teams get here? Were they teams that were leading the pack during the season and fell off or teams that were lagging behind that sped up? The laws of averages are mostly true in this case. Of the teams that finished in 7th, they averaged 6.7 wins after 12 games and finished with an average of 9.1, proving that they were slightly better than average (2.4 wins) over the last 4 games. Interestingly, three of the four worst teams after 12 weeks (two teams were 4-8 and two were 5-7) finished the season on 4-game winning streaks. All four of the 7th place finishers that were the worst after 12 games only missed out on the playoffs because of tiebreakers. Another outlier is the top team after 12 weeks (the 9-3 Bucs of 2008) were the only team to lose all 4 games down the stretch and still finish in 7th.

On the whole, the trends we have seen throughout this analysis remain true: most of the teams that finish in 7th were middle-of-the-pack teams that were battling for position after 12 games and only barely missed out on the postseason. Most of the 7th place finishers did close out the season strong with at least 1 win and 10 of them finished with wins in at least the last two weeks, which supports the theory that 7th place teams could pose a threat in the playoffs and would not simply be a stomping mat for the #2 seeds. In fact, as we remember from Part I of this series, many of the potential #7 seeds had scored victories over the 2-seeds during the season. Another trend to note here is that adding a 3rd wild card team does allow for more "forgiveness" of early-season set-backs. As we saw with the 4 teams that were either 4-8 or 5-7, a strong finish to the season would have a greater probability of propelling a poor-starting team into the postseason. Whether this is good or bad for football is debatable, but it may to an extent lessen the blow that some teams suffer when a significant play-maker is injured. For instance, despite Jay Cutler's injuries which led to mid-season losing streaks, the Bears would have still made the playoffs in 2011 and 2012.

Speaking of the Bears, based on past data, the NFC North is the division that would have benefitted the most from having 7 teams in the playoffs. These findings don't really have much bearing on the future, as (at least in theory) every team has an equal chance of finishing 7th at the beginning of any given season. Looking at the historical data, the addition of a 7th team would not be a case of the rich getting richer. The 5 teams with the most playoff appearances (Indianapolis, New England, Green Bay, Philadelphia, Seattle) only accounted for 2 of the 24 7th place teams. The biggest benefit was seen by the "middle of the road" teams such as Minnesota and New Orleans who would move from the middle of the pack in terms of playoff appearances into the top 10. Denver and Pittsburgh would move from 8th and 6th in the league in playoff appearances into the top 5. Chicago would move from 21st into the top half. Interestingly, as you may remember from my post on the historical breakdown yesterday, two of these squads (Minnesota and Chicago) fired coaches after finishing in 7th place. One would have to wonder if Mike Tice and Lovie Smith would have kept their jobs if they had made the playoffs as the 3rd Wild Card.

The final point of analysis, which is a little harder to gauge, is the potential impact on #2 seeds. Under the current alignment, 2-seeds get byes to the second round. With the new alignment they would be forced to play an opening round game against the 7th place team. On one hand this could be beneficial to #2 seeds as it could allow them to get on a run and build momentum towards a Super Bowl run. On the other hand, it provides another opportunity for injury and could derail their Super Bowl hopes even sooner. 2 seeds have been moderately successful in the playoffs since the realignment, accounting for 5 of the 22 teams to have reached the Super Bowl. By contrast, #3 and #4 seeds (who have had to play an extra game) have only combined for 5 Super Bowl appearances. Wild Card teams have only reached the Super Bowl 3 times, but they have won all of their appearances. In fact, teams that have to play on Wild Card Weekend that do reach the Super Bowl have been incredibly successful, going 6-2. One other interesting note from the playoff data is that from 2002-2012 there was been no difference between #3 and #4 seeds in the first round, with both seeds going 13-9. Of the 9 6-seeds that have managed to advance past the first round, they have held their own against top-seeded teams, going 5-4 while #5 seeds went only 3-6. Both #3 and #4 seeds that made it out of the first round suffered similar fates in the second round, with both seeds posting 4-9 records. The most Super Bowls have actually been won by #2 seeds (3) though they have reached the Super Bowl less often than top seeds, who have accounted for 9 of the 22 Super Bowl teams (though #1 seeds are just 2-9 in Super Bowls since 2002). As the third chart shows, the danger of playing an extra game is that teams have a lower win percentage in their second and third playoff games (though the reason that the win percentage in second games is so low is mostly due to the poor records of teams from Wild Card Weekend in the divisional round (16-28) where they faced teams that had a bye. This is were 2-seeds will lose out the most. 2-seeds have been the best performing group in their opening playoff games (a 15-7 record as opposed to 13-9 marks by the other division winners and 9-13 marks by Wild Card teams), yet they will lose their advantage of having the extra bye week under the new system.


So what conclusions can we draw from expanding the playoff field? First, that the field will not necessarily be muddled by worse teams making the postseason. As discussed, all of the 7th place teams in the last 12 years have had at least 8 wins and there have been slightly more 7th place teams with double-digit wins (7) as there have been with .500 records (6). Secondly, these are teams that are hot entering the postseason. All but 4 of them won at least two of their final 4 games and all but 6 ended the season with a winning streak. Third, it is hard to say that most of these teams are "undeserving" of making the playoffs as 11 of the 24 missed out because of a tiebreaker and all but 4 finished within a game of the #6 seed. On the whole, 14 of the 24 7th place finishers ended the season there on their own and did not earn the spot via tiebreaker.

No comments: