Stats Analysis - as of Oct. 8th

Why Details Matter

Wait, that can't be right...

This fall has seen one of the most confounding statistical train wrecks I've ever seen, let alone been a party to.

Haverford College has seen it's men's soccer program play it's way to a solid 7-3 record, which is nice.  But at 2-2 in Centennial Conference play, the Fords sit precariously at 6th place of ten, with only the top 5 advancing to the conference tournament (4 & 5 play off mid-week to see who advances as the 4th and final semi-finalist on the weekend).

While the offense is doing alright (2.2 goals per game) it is on the other end that the Fords are falling short, furthermore, they fall short in an astounding fashion:

This is the Fords' save percentage over the 10 games; 20 saves, 13 goals.  60 % save percentage. 

The save percentage chart shows something amazing...look at the number of saves (first column of numbers following the school name) and see that Haverford is one of only two schools making fewer than 30 saves on the year.  Add that to the 13 goals conceded and the Fords have allowed only 33 shots on goal in 10 games.  That is impressive in and of itself.  But the save percentage of those shots is 385th in the nation, just 7 percentage points from dead last at 398th.  Other than Whittier of the less-than-powerful SCIAC, every other team in this particular basement has given up minimally double the number of goals of Haverford.

The Fords back four can take pride in this, to the extent that in 10 games, the GK has only made 2 saves per game...good enough for the second lowest saves per game average in the NCAA.  On the other hand, those two saves per game only average out to be 60% of all the shots on frame.  By contrast, the number 2 team in the nation (Rutgers-Camden) is just under 84%...the bottom half begins at 75 and a half % on the nose.  To crack the top 25 in save percentage a team needs an 86 and a half % rate.

Refer back to the graphic of the Centennial-only stats at the top of the article, and note that of the 10 shots on target, 7 have found the back of the net.  That's a terrible rate...but think about it, 10 shots on target in 4 games??

Read it carefully and it looks like this:

Shots On Goal Allowed by team
HAVERFORD                --     10
Dickinson                         --     19 (2 GKs combined)
Gettysburg                       --      28
Hopkins                           --      27
Swarthmore                     --      15
Ursinius                           --       24
McDaniel                         --      18
Franklin & Marshall         --      14
Washington Coll.             --      21
Muhlenberg                     --      14

The next closest two teams to Haverford have allowed 29% more shots than the Fords.  Three teams have given up 2.4 TIMES as many shots or more!  Talk about an outlier.  But the Fords save percentage is 34% worse than the second-worst defense.

The Fords will turn to rookie net minder, Sam Miller, this week in the wake of a training ground injury to Nick Kahn...ready or not, Miller will need to find a way to increase the save percentage immediately in Nick's absence.  It's unlikely the defense can do much more, given the incredibly low number of shots allowed.  That would be essentially asking for an improvement on perfect!

But there is clearly a detail being overlooked which leads to these goals despite otherwise fine defensive performances.  What's scary to the Fords' staff is that we can't know for sure that we identify the correct detail...good luck could mask our oversight temporarily, only to abandon us later in the season and leave us vulnerable to the same type of goals.  This is the stuff that keeps coaches up nights.