Does defense actually win championships?

We take a look at the numbers to determine the validity of this long-held claim.

Defense Wins Championships. If you’re a sports fan, I’m sure you’ve heard this phrase a billion times. No matter the sport, this always floats around in conversation, especially when a relatively good defensive team is brought up in conversation. It definitely sounds true, but is there any proof? Have you ever seen a statistic or comparison between defense and success that proves this theory? I certainly haven’t, but still, the phrase looms large over the sports world, and is essentially regarded as fact and nomenclature instead of an ideal. So, being the sports skeptic I am, I decided to attempt to prove it. I combined my science brain and my sports brain and spent hours gathering records over the past decade or so, to ask the question: does defense actually win championships?

I looked at the three major sports in America: baseball, football, and basketball, and I looked at professional major leagues because those are the best of the best athletes. Sorry hockey fans, you were a close fourth. My first question was simply: does defense correlate with wins? To test this, I looked at the defensive rankings of the sports major defensive category (football, yards per game [YPG]; baseball, earned runs allowed [ERA]; and basketball, points allowed per game [PPG]). I plotted out how these rankings related to wins in each respective sport. I used data in the NBA and NFL back until the most recent expansion teams were added to each sport (2001 NFL, 2005 NBA) and for the MLB, I used every year of the current century.

Starting with the NFL:

hm-graph-1

First, I need to note that Series 15 is the 2015 NFL season and Series 16 is a reference of

hm-graph-2winning exactly half of your games (8). Therefore, points above this reference show teams with a winning record, and teams below have a losing record. Second, this is a jumbled mess. You can somewhat make out the trend that the lower the defensive rank, the farther down the values are, but this is still hard to interpret. To help with this, I added some linear trendlines of this data to see if the correlation could be made more clear:

The correlation is much more clear here. The black line represents a “perfect” representation of this phrase, where the best defense wins all the games, and the worst wins none of them, again as a point of reference (a 1:1 correlation). As you can see, the trends in the data are in the direction that would support this theory. Even what seems to be the lowest trend of the data in 2011 (yellow) still has a positive R 2 value (0.011) which indicates that there is still a positive correlation between defensive rank and wins.

I found very similar results in the NBA and MLB correlations respectively between defensive rankings and wins:

hm-graph-3

hm-graph-4

(Lowest green line represents NBA labor strike year in which only 66 games were played)

hm-graph-5

hm-graph-6

There is a clear correlation between defense and wins in any given season from this data. But who cares? If the Carolina Panthers or Golden State Warriors taught us anything last year, it’s that you can win almost every game and still come up short when it matters in the championships. So my next test was to take these same defensive rankings year by year, and then compare only the rankings of winners and losers of the championship game that year, to see if the better ranked defensive team had the advantage where it mattered.

This is what I found for the Superbowl (also to note, the x-axis represents Superbowl years, with 1 being the 2016 Superbowl, 2 being the 2015 Superbowl, etc.):

hm-graph-7

We would expect if defense won championships, for the winners in orange to have a higher defensive rank, and therefore have lower points on the graph (red circles represent the unexpected result). However, this only happens eight times over the last 14 years. That’s approximately 57% of the time that the better defensive team won the championship game, and this is not frequent enough to make the conclusion that the better defense will win the championship.

I found similar results yet again with the NBA and MLB respectively:

hm-graph-8

hm-graph-9

With this data combined, the better defensive ranked team only wins the championship game or series approximately 49 percent. That’s obviously less than half of the time. This is obviously not what is expected, probably because the theory that defense wins championships has been accepted far too easily.

But I know sports. I know that a team probably isn’t going to play exactly the same offensively or defensively in the playoffs as they did in the regular season. Maybe teams and/o players believe this theory too, and they decide to step it up and play a better defensive game when it really counts. I wanted to try to account for this using NBA and MLB defensive statistics throughout the finals or world series, because it is a series of games, as opposed to one game in the NFL. However, the MLB statistics I needed weren’t easily available, so I was only able to look at the NBA finals of the same time span I have been throughout my study.

hm-graph-10

What I found was that over this span of the last 12 years, teams that were able to hold their opponent to a lower field goal percentage (percentage of shots made that were attempted [FG%]) won the finals all but one time (2010). And teams that were able to hold their opponent to the fewest PPG won the finals every year but two times (2013 and 2010). To be sure, I also did go check, and every single one of these values, FG% and PPG, was lower for the losers than their regular season averages. This data shows that in the NBA finals over the last dozen years, the team that has been able to perform better defensively has consistently won a high majority of the series.

Does defense win championships? This data shows it is positively correlated with winning in the regular season. However, these successful defenses prove to be not as much of a factor as anticipated when it comes to winning championships. The data of defensive rankings related to championship wins is inconclusive, but the NBA finals defensive performance data proves that there may be more to this theory than rankings suggest. Winning a championship is the most difficult thing to do in every sport, and is also the most important. There are many factors required into having a championship team. Defense is no small part, but don’t necessarily look at it to be the single make-or-break factor of any team or player being successful.

Jake Marlay

Jake is a senior biology major who likes sports and served as the Sports Editor for The Monitor from the Spring of 2017 to the Spring of 2018.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.