January 1, 2011
In The Crease
The Consistency of Marc-Andre Fleury

by Philip Myrland

Ask a scout to describe Marc-Andre Fleury and one of the first words out of their mouth is likely to be "athletic" or "athleticism". Fleury is lightning quick, moves well around his crease, and has a flair for making the dramatic save, often with his glove hand as when he stopped Alexander Ovechkin on a breakaway in Game 7 of the fabled Pittsburgh-Washington playoff series in 2009. Any casual hockey fan can instantly recall a number of other big saves from the man whose teammates call him Flower.

However, while the ability to make difficult saves is certainly a significant part of being a quality goaltender, there is more to it than that. Goalies have to make the easy saves too, as well as contribute to their team in others ways, such as controlling rebounds and playing the puck. Both of those have at times been problematic skills for Fleury, as was highlighted in HBO's excellent 24/7 series when Pittsburgh's own coaching staff was caught on camera pointing to Fleury as the weak link in the Penguins' defensive zone puck retrieval during one particular game.

This is partly why the one of the next few words from that same scout is likely to include "consistent" or "consistency". Despite his natural talent, the perception is that Marc-Andre Fleury struggles with inconsistency and runs hot and cold. Some say that on his best game he is the rival of any netminder in the league, and if he can just improve his consistency from night to night then he will be among the game's elite.

This season certainly seems to be good evidence of inconsistent play from Fleury. In his first 11 starts, Fleury was 3-6-1 with a 3.05 GAA and an .873 save percentage, and it seemed like his starting job was in jeopardy. In his next 11 starts, Fleury posted a 1.62 GAA and a .945 save percentage. With numbers like that, it's not surprising that the stretch coincided with the Penguins' long winning streak, with Fleury and his teammates winning in all 11 of those starts.

The truth is, however, that all NHL goalies are capable of getting hot, and all NHL goalies are capable of stonewalling the opposition on their best day. Crazy hot or cold streaks are very common, and afflict both great and mediocre alike. What separates the best from the rest is the ability to not just have more dominant outings, but also bringing a higher level of performance in any typical night at work, not just when they are really feeling it and the puck is bouncing their way. Those games where it seems to be one masked man keeping the whole world at bay tend to be remembered, especially if those signature games come under the spotlight of playoff or international play. In the long run, though, the best goalies are those that make the most saves, period, not the ones who perhaps have the most memorable single game performances.

Traditionally, one of the ways of rating the ability of a goaltender to dominate a single game is by looking at his shutout totals. This is an imperfect method since shutouts are impacted by the defensive play in front of a goalie, and stopping 44 of 45 shots is often a better result than turning aside 16 shots against a weak opponent. In addition, Fleury is potentially disadvantaged by playing for a team that has had a tendency to seek out more offense even at the expense of leaving their own netminder vulnerable.

That said, Fleury has faced a relatively average level of shots against per game since the Lockout, and despite that, his shutout results are very average. Fleury ranks just 17th in the league since the Lockout with 16 shutouts. Considering only seven other goalies played more games, that is not a terribly great record, and does not indicate a goaltender who routinely keeps the opposition off the scoreboard on his best nights.

A more detailed analysis of Fleury's game-to-game variance requires comparing his results against the game-to-game numbers of his peers. The control sample in this study consists of the 15 goalies other than Fleury with the highest games played since 2006-07. All of them have over 200 starts in that period. The sample includes both regular season and playoff results to increase the sample size for Fleury. How does he compare?

Marc-Andre Fleury vs. his peers since 2006-2007, Goals Against

GA	Field	Fleury
0	7.6%	7.6%
1	19.4%	16.9%
2	26.5%	23.3%
3	24.1%	28.5%
4	14.9%	16.1%
5	5.3%	5.6%
6+	2.2%	2.0%

The main difference is that Fleury is less likely to allow 1 or 2 Goals Against than the larger group of starters, and more likely to concede 3 or 4 Goals Against. That suggests that Fleury simply has a higher average, rather than a significantly different distribution. Fleury was slightly more likely to get pulled (he played 40 minutes or less in 6.4% of his starts, compared to 5.9% for the other goalies), but he was not any more likely to get shelled for a high number of goals against.

A look at save percentage tells a similar story:

Marc-Andre Fleury vs. his peers since 2006-2007, save percentage

Range		Field	Fleury
.000-.750	3.1%	2.4%
.751-.775	1.3%	1.6%
.776-.799	1.7%	0.8%
.800-.824	3.6%	2.8%
.825-.849	4.9%	6.8%
.850-.874	8.1%	8.8%
.875-.899	13.0%	15.3%
.900-.924	18.9%	21.3%
.925-.949	19.1%	16.5%
.950-.974	17.8%	15.3%
.975-1.000	8.4%	8.4%

Again, Fleury does not have more a higher rate of awful starts than the league's most dependable starters. Neither does he have a higher rate of outstanding games. What he has is a distribution that peaks slightly earlier than the rest, suggesting that the main difference is not his consistency, but a slightly lower level of talent than the others. This becomes apparent when the distributions are compared graphically:

It is very easy to overrate goalies that are athletic and make exciting, highlight reel saves. That likely accounts for why goalies like Grant Fuhr and Mike Richter are remembered much more fondly than their save percentage records would suggest they deserve. Goaltending is about a lot more than athleticism. It is possible that the last two months indicate that Marc-Andre Fleury is finally putting his game together and making the leap to elite goalie status, but it is also possible that he is riding a temporary hot streak.

Whatever the future holds, Fleury's historical numbers show that the narrative of his inconsistency is not correct. The difference between Fleury and the league's best starting goalies has not been the frequency of great and bad games, but instead his overall level of performance.

Philip Myrland is an author of Hockey Prospectus. You can contact Philip by clicking here or click here to see Philip's other articles.