In their last t̶w̶o̶ t̶h̶r̶e̶e̶ four league games, Pacific FC have started veteran Canadian goalkeeper Sean Melvin. But if they want to optimize their chances of winning they should instead start Whitecaps loanee Max Anchor. 

This article will be short and fact-based. However, we should first address the elephant in the room. I am partisan on this issue. I am a Whitecaps fan and I want Whitecaps players who go on loan to play games in order to give them the best possible chance of becoming first-team contributors. But my partisan interest is not the reason I think that Max Anchor is the better goalkeeper. Rather, my partisan interest is the reason I am trying to use whatever influence The Third Sub has on the BC soccer community to raise the salience of this issue. 

Post-Shot Expected Goals

At the end of the day, the most important part of goalkeeping is stopping the ball. There are other elements to the position of course but it takes a lot of crosses claimed and astute distribution to make up for a goalkeeper who is a below-average shot-stopper. The best way to measure a goalkeeper’s shot-stopping is through post-shot expected goals. Normally when you see the xG stat it is “pre-shot.” That is to say that the data analysts essentially pause at the moment the ball is about to leave a player’s foot, look at the situation, determine how many goals were scored from that sort of situation in the past, and then assign the shot a value. But this way of measuring xG is less useful for goalkeepers because a rocket into the top corner from 20 yards is a lot harder to stop than a shot from 6 yards where the striker kicked the ground instead of the ball. So, for goalkeepers, we have post-shot expected goals where more information about where the shot is placed is included in the calculation. 

Max Anchor is Pretty Good

Max Anchor’s history with post-shot expected goals is quite strong. He has played significant minutes in three MLS Next Pro seasons and saw steady year-on-year improvement in each of them. This culminated in the 2024 season where he stopped 0.11 goals per game more than the average keeper, or about three extra goals prevented over the course of a 30-game season. This isn’t world-beating stuff but given Anchor’s age and his continued improvement it does suggest a trajectory that is arching towards a career as at least a decent MLS goalkeeper.

Now, I think CPL is a slightly higher level than MLSNP. A league largely composed of adults trying to put food on the table is always going to have a higher level of intensity than a reserve league with an average age of about 21 (that’s why a year in the CPL was a logical next step for Anchor). But not every CPL player who has gone to MLSNP has dominated it and some players who washed out of MLSNP have gone on to be reasonably successful CPL players. So they are clearly leagues that are in the same neighbourhood in terms of quality. It is therefore reasonable to assume that in a big enough sample size, Anchor would also be an above-average shot-stopper in CPL. 

Sean Melvin is Not Very Good 

Sean Melvin has not been an above-average shot-stopper a single time in his professional career. In fact, his numbers have been quite bad most of the time. In the two CPL seasons Melvin has played significant minutes he has been -0.3/90 and -0.46/90 respectively. In other words, Melvin is conceding an extra goal compared to the average keeper every 2-3 games in CPL. His average would see him concede 11 more goals than the average keeper over the course of a 30-game season*. 

Now, this underperformance does not necessarily always manifest itself as big and obvious errors. It might be that Melvin concedes on shots you would expect the keeper to save 60% of the time only 30% of the time or things of that nature. Conversely, Anchor’s over-performance does not mean that he never makes mistakes or that there are never shots he could do better on. Indeed, in the first three games where Anchor played, I think there were a couple of goals he could have given himself a better chance of stopping. But the numbers could not be more clear, if you want to give yourself the best chance of winning, Anchor is the clear choice.

*I received a reply on Reddit, which, while not invalidating my overall argument, did make some good points. So I wanted to add this addendum, which clarifies this point slightly. If we look at the entirety of Melvin’s career (a much bigger sample size than his 20 CPL appearances), we see that 11 goals might be a slightly dramatic number.  If we include his 56 USL Championship appearances in our average, then that would ‘only’ see Melvin costing his team 5 or 6 goals compared to an average keeper. The Whitecaps banished Thomas Hasal to the shadow realm for putting up similar numbers in MLS. Here’s the thing though, that still sucks! Football is a low scoring game and teams spend a lot of money to improve by 5 or 6 goals. Plus, this average includes Melvin’s physical prime. Between natural decline due to age and generally playing few competitive games in his late 20s, I would not be surprised at all if Melvin ends up being worse than his previous career average this season.