Swish Appeal

For those of us that have been following the WNBA closely during the 2009 season, it would probably sound odd – if not absolutely ridiculous – for someone to suggest that the Phoenix Mercury’s strategy of wearing opponents down doesn’t actually help them win.

The other day I received an email suggesting exactly that: statistically, there is little difference in how the Mercury performed relative to their opponents in the 4th quarters and overtimes compared with the first three quarters of games. I received the following statistics along with that email:

Of course, this information is contrary to what we’ve heard about the Mercury for most of the season.

So if Phoenix Mercury coach Corey Gaines has spoken repeatedly about the strategy of pushing the pace and wearing their opponents down in order to win, then what's going on?

The standard theory goes that if the Mercury push the tempo for 40 minutes and wear their opponents down, they’ll be able to outperform their opponents down the stretch. Win or lose, Gaines has stuck to the strategy, as exemplified by the similarity of his comments after both their July 18th win and September 1st loss against the Detroit Shock.

After the September 1st loss, Gaines said the following:

Phoenix (20-10) got 25 points from Cappie Pondexter and 20 from Diana Taurasi, who missed a potential winning 3-pointer as time expired.

"They played well and we just didn't get that last shot in," Phoenix coach Corey Gaines said. "We tried to push the pace and wear them down, but they hit some big shots and we didn't."

After the July 18th win against the Shock -- a game MVP Diana Taurasi missed due to suspension -- Gaines said something similar and York elaborated on the strategy a bit.

With as fast as the Mercury play, it’s no surprise how physically conditioned they are and if they need to, any given player can contribute more minutes than planned. In overtime, it’s a perfect time to wear teams down and push up the tempo even more; that’s precisely what the Mercury did in the overtime period scoring 14 points in 5 minutes and holding Detroit to just 7.  "I told them it was to our advantage," Gaines said about heading into OT. "I told them to keep pushing the ball and to go faster. We outscored them 14-7 in OT so we were able to do that."

So of course, entering the WNBA Finals, the dominant assumption was that the Indiana Fever would also be at a disadvantage if they tried to run with the Mercury and would have to slow the pace in order to have a chance at winning.

And in Games 1 and 2 of the Finals that has definitely been challenged – the Fever have proven that they can both run with the Mercury and win.

Therefore, it is reasonable to ask questions about two assumptions in light of this evidence: does the Mercury’s pace actually wear down their opponents? (And, if so, what is the effect?) How does it affect the Fever?

Assumption 1: the Mercury wear opponents down and then outperform them at the end of games.

The Mercury's regular season statistics seem to indicate that the Mercury’s pace had little, if any, affect on their opponents.

However, when you see opponents leaning over with their hands on their knees or tugging on their shorts in the fourth quarter, it would seem intuitive that there is some effect on performance.

So first, I wanted to know what the pace was in those time intervals. Second, I wanted to see if we could ferret out some differences by looking at more advanced statistics.

As it turns out, while there is no significant difference in points scored, there was some difference in their Four Factors numbers during the season, specifically:

• Free throw rate (which is considered by some a proxy for a team's ability to put pressure on the defense)
• Offensive rebounding percentage (the percentage of available offensive rebounds that a team secures)
• Assisted field goal percentage (not one of the four factors, but I've found it to be a reasonable proxy for how well a team moves the ball to set up scoring opportunities)

In addition, measuring production using David Sparks' Model Estimated Value formula, it is clear that there is a decrease in overall production from teams as well.

Looking at the differentials between the final period(s) and the first three, we can see that while Mercury opponents increased their free throw rate in the 4th quarter & OT, the differential between the Mercury’s assisted field goals as well as offensive rebounding percentage increased.

While it’s an interesting observation, it’s also worth noting that this really had no bearing on the scoring differential, as we saw from the original table of stats about the Mercury's regular season performance.

Regardless, what we're most concerned about now is the FInals and what we’ve seen in the Finals is something even further from the theory that the Mercury benefit by wearing opponents down.

Assumption 2: The Fever cannot win if they run with the Mercury.

In the Finals the Mercury keep saying they want to increase the pace as well, but what is actually happening is that the Mercury have been best when the Pace drops much more considerably than it did in the regular season.

(To see the full table of regular season and Finals statistics, click here)

Similar to the regular season, the Mercury have gained a considerable advantage in offensive rebounding at the end of games in the Finals. Also in keeping with the regular season, the Fever's overall production in the final periods has dropped.

But the similarities end there.

Although the differentials in free throw rate were almost even in Game 1, the Fever's free throw rate skyrocketed in Game 2 while the Mercury's dropped considerably.

However, the biggest difference between the Finals and the regular season is the differential in assisted field goal percentage, which has dropped at a much higher rate for the Mercury than the Fever in both games. In Game 2, the Mercury's assisted field goal percentage dropped a little, while the Fever had a significant increase in their assisted field goals (which is partially because they had 3 assists on 4 shots).

Most importantly though, what we saw during Game 2 is that the Fever were most productive and scored the most points in quarters in which the pace was higher (similar to Game 1), seeming to imply that it is the Fever – not the Mercury – that benefits when the tempo increases.

Albeit a small sample, this seems to challenge the notion that the Mercury can wear down the Fever by picking up the pace in addition to suggesting that this series will not come down to a matter of merely increasing or decreasing the pace.

"A lot of people question whether we can get up and down with them," said Fever forward Tamika Catchings. "We do run. I think the key for us is they want to run all the time and we have to learn when to slow it down."

So where does all this leave us?

Obviously, we cannot use a sample of two games to say that the Mercury's strategy of running all the time will not work against the Fever. However, we can say that the Mercury will have to do more than rely on a high pace to defeat the Fever.

Differentials aside, the biggest difference for the Mercury between the two games were rebounding and free throw rate. If the Mercury can use their quickness to crash the offensive glass again and drive to get to the free throw line, they might be more effective. And of course, it seems unlikely that both Taurasi and guard Cappie Pondexter will have off games again.

Then again, it's hard to come to any statistical conclusions after two games.

Ultimately, this series might not come down to who is able to run with whom, but figuring out when to pick up the pace and when not to.

Statistics cannot really describe situational decisions, so we’ll have to wait to see how both teams respond to the first two games in Game 3.

Transition Points:

• My subjective opinion: It's not about wearing other teams down as much as it is about having players like Taurasi and Pondexter who are able to rise to the occasion in ways that other players just can't. In an effort to support that point statistically, I took a look at the "primetime performers" statistics on the Swanny's Stats blog, but couldn't immediately identify anything that might help them late in games above and beyond just being good. Yes, that is my theory -- this is just a good team.
• Whether these particular differentials are good or bad, depends on the opponent – for a team whose success is predicated on ball movement or offensive rebounding (and the associated second-chance points) this may matter because the Mercury might be able to take them out of their game. Conversely, a team that doesn’t shoot free throws well probably won’t be able to exploit the increased free throw rate. So there is an effect, but a varying effect depending on the opponent...maybe.
• An interesting hypothetical addendum with these numbers: if I had infinite time, it might be interesting to figure out if teams perform worse at the end of games against the Mercury compared to other teams.