SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.
Also, our existing PFR blog rss feed will be redirected to the new site's feed.
Pro-Football-Reference.com » Sports Reference
For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.
ShareThisConsistent franchises
In the comments to Friday's post, PackerNation observed:
Look at the Jets….they’re 6-10 or 10-6 or somewhere in between every year
And indeed, they have been in that range for 8 of the past 10 seasons. Apologies to PackerNation for lifting his quote out of its context, but I wasn't interested as much by the context as by the the fact that it gave me an idea. Namely, which franchises have been the most and least consistent over the past several seasons?
So I decided to calculate each teams' standard deviation of seasonal wins since 1978. As is typical, I made a guess about what the answer was going to be before I ran the query. As is very from from typical, my guesses were pretty close.
I immediately thought the 49ers would turn out to the be least consistent team; they've been way up and way back down a few times over the last few decades. They've got as many Lombardi trophies and less-than-5-win-seasons combined as they have between-6-and-10-win seasons. They do indeed have the highest standard deviation: 3.96 wins. Other inconsistent franchises include the Colts, Titans, Patriots, and Bears.
You might think the 49ers have the highest average number of wins during the period in question, but they don't. In fact, there are three teams with higher winning percentages and, oddly enough, all three of those teams rank among the most consistent franchises. They are the Dolphins, Broncos, and Steelers. The Dolphins, for what it's worth, were my guess at most consistent.
But the Dolphins have not, in fact, been the most consistent. No, the most consistent team is also the worst (not counting the Texans). It's the Cardinals, of course, and in my mind this obliterates any argument that Lion or Bengal fans think they might have in a franchise futility contest.
I'll close with a fun chart:
========== Number of N-win seasons ============
AvWins StDev 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
==================+===============================================
crd 5.91 1.69 | 4 1 6 3 9 5 1
htx 4.80 1.72 | 1 1 1 1 1
mia 9.53 2.12 | 1 5 4 4 7 5 2 1
sea 8.02 2.14 | 1 1 3 9 4 5 4 1 1
min 8.48 2.19 | 1 3 3 10 4 3 3 1 1
den 9.65 2.41 | 1 3 2 5 5 4 5 1 1 1 1
pit 9.44 2.42 | 1 1 1 2 6 2 7 2 3 3 1
kan 8.37 2.46 | 3 2 4 6 3 4 3 2 2
det 6.41 2.53 | 1 2 5 2 4 2 7 2 2 2
phi 8.80 2.60 | 1 3 5 7 1 1 5 2 3 1
rav 8.32 2.63 | 1 1 2 1 1 2 2 1
gnb 8.37 2.65 | 2 2 2 4 5 6 3 1 4
nyj 7.52 2.66 | 1 3 4 4 5 2 4 4 1 1
atl 6.90 2.69 | 1 1 1 2 3 2 6 3 4 3 3
nor 7.31 2.70 | 1 1 1 3 2 6 7 2 1 1 3 1
cle 7.05 2.72 | 1 3 2 3 2 4 2 4 2 2 1
nyg 8.22 2.74 | 1 1 2 3 3 3 3 3 6 1 3
tam 6.54 2.80 | 1 2 3 3 1 2 5 6 2 1 3
was 8.64 2.88 | 3 2 1 6 3 4 2 4 2 1 1
rai 8.40 2.88 | 1 4 3 2 4 5 4 2 3 1
car 7.50 2.90 | 1 2 2 5 1 1
jax 8.50 2.93 | 1 1 2 2 1 1 2 1 1
cin 6.68 2.94 | 3 1 1 1 6 4 2 1 5 4 1
sdg 7.77 2.97 | 1 2 4 1 4 5 1 4 2 4 1
ram 8.00 3.01 | 1 1 2 3 4 2 2 2 5 3 3 1
buf 8.05 3.05 | 2 2 3 4 2 3 5 2 2 1 1 2
dal 8.87 3.05 | 1 7 3 4 3 1 3 2 3 1 1
chi 8.38 3.20 | 1 1 2 2 3 2 2 1 5 3 4 3
nwe 8.46 3.24 | 2 1 5 4 7 2 2 3 2 1
oti 8.01 3.30 | 2 2 4 4 2 4 3 3 1 1 3
clt 7.15 3.62 | 1 1 3 3 4 2 3 1 3 2 3 1 2
sfo 9.40 3.96 | 1 4 3 3 2 6 2 2 1 2 3
Fine print: a "10-win-season" is actually one in which the team's winning percentage, when multiplied by 16, gets rounded to 10. Likewise, the AvWins and StDev columns in the chart above are actually 16*AvWinPct and 16*StDevOfWinPct, respectively. All this junk is only relevant for strike years and seasons that have ties.
This entry was posted on Monday, March 26th, 2007 at 4:01 am and is filed under General, History. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
Put an asterisk next to the Bears' 13 win season in 2001.
As a Dolphin fan, I expected the Dolphins to be near the top (I hadn't considered the idea of consistently bad teams like Arizona). I've long felt the pain of rooting for good teams, but teams that are never good enough to go to the Super Bowl.
I was surprised to see Seattle up there, but then I remembered their run of back-to-back 8-8 seasons. Looking on their page, I see that from 1995 to 1999, the Seahawks had a sweet run of 8,7,8,8,9 wins. Seeing that 26 of their 29 seasons had between 6 and 10 wins is awesome.
I'd like to take this time to point out that in the last ten years, the Jets have had two losing seasons. Only Green Bay, Denver and New England have fewer (one each). New England, Miami and Denver all have eight winning seasons in the past ten years, and the Jets are behind them with seven.
I'm not sure about the methodology here. I would consider an "inconsistent" team as one whose record varies wildly from year-to-year. Using your method, a team's record may change only slightly--if at all--from any given year to the next, but cyclic cumulative effects over time could give it a high stddev from the mean.
Perhaps a better measure would involve the correlation between a team's record in one year and its record in the year immediately prior (or after).
Kurt, I'm taking a more macro view of consistency.
In my mind, the 49ers are inconsistent because they've been both very good and very bad several times during the given period. I am not particularly interested in the chronology and whether their goods were bunched with goods and bads with bads. At least I wasn't interested in it for this post. It might indeed make an interesting post, though.
I think perhaps the problem may be in the subjectivity of the word. While the general concept of "consistency" is well-understood, the technical question of what, then, meets the criteria set by that concept is much more debatable. I would not consider a team that, say, had a string of great years in the late 70's-80's, then had a few years of (steady) decline and then has been total crap since to be "inconsistent" in any meaningful sense--even when they were changing, the change itself was inconsistent.
It's kind of like Popper's assessment of Plato's claim that the "wisest" and "best" should rule--that's all well and good, and goes without saying, but it really doesn't help you decide anything.
Perhaps it might be an interesting experiment to use several statistical methods, and then see which one correlates best with peoples' individual subjective evaluations of teams' consistency over time.
Indeed, Kurt.
The problem is that I run some numbers, find them interesting, write them up, and only then do I say to myself, "ah crud, I gotta come up with some sort of title for this thing." By then I'm sick of writing, so I just bung down the first thing that pops into my head.
This *does* measure what I wanted to measure, but I'll acknowledge that consistency might not be the best word for it.
I'm pleased to see you're upgrading your source material...........
My point in my observation was that consistent mediocrity is not a virtue. Would Miami have traded all that consistency for a Super Bowl title or two? Would the highly inconsistent 49ers rather be consistently mediocre or are they happy with their Lombardi trophies?
I think that franchises, like the Jets, who are mired in consistent mediocrity might benefit from taking more risks in the draft, whether it be trading up or whatever.
Those consistent 6-10 win seasons leave you drafting in the middle of the pack and you don't find the special players there as often as you do in the top 10 or top 5. I just think that at some point, management has to say "Screw it. We're going to try to make something happen here even if it blows up in our face."
Here's an alternative way to measure consistency, that better matches what Kurt was suggesting.
First, for strike years and years with ties, do the same thing as you did here, multiplying winning percentage by 16 to get a win total.
Now, what you measure is win difference over each 2 year period, and then average them. So take the 1999-2000 Colts, who regressed from 13-3 to 10-6. That's a win difference of 3 for 1999-2000. (Note that you're taking absolute value, so there can't be negative changes in wins.) For 2000-2001 and 2001-2002, they have win differences of 4 both times, going from 10-6 to 6-10 to 10-6. So from 1999-2002, the Colts have an average win difference of 3.33.
I think this measures consistency pretty well. Let's measure the two Super Bowl teams, the Colts and the Bears, and how they've performed since 1999. The general perception is that the Colts have been a consistently great team in that era. The Bears, however, were terrible, then they had the fluke 13-3 season, then they were terrible again, and then they made a dramatic return to excellence in 2005. They're not considered a consistently great team.
Colts' win differences: 3, 4, 4, 2, 0, 2, 2.
Average 2.4
Bears' win differences: 1, 8, 9, 3, 2, 6, 2.
Average 4.4
The most consistent team since 1999 has to be Arizona: 3, 4, 2, 1, 2, 1, 0
Average 1.9
In a later post, maybe I'll describe a slightly more complicated system based on the same principles.
Chase, you're a homer
I am happier as a Dolphin fan from 1990-now than I would have been if I were a Falcons fan over that same period. For most of the 1990's and the first few years of the 2000's, I had high hopes about the Dolphins maybe getting to the Championship game, and missing the playoffs was a disappointment. The Falcons did play in a Super Bowl, and lost in another NFC Championship game, but most of that time were a bad team with low expectations.
dude why do I feel like I need to be Neo from the Matrix to read that chart?
Doug, I have decided that your team is the Minnesota of our league. GLS.
So now does this make Jake Plummer the greatest QB ever?