Your MLS Ref Cheat Sheet, or My Lengthy Apology to Mark Geiger

Mr. Geiger, I'm sorry.  I hope that this article will start to mend our one-way relationship, which heretofore has been pretty rocky.  For the past few years I've gone around dogging you as a red card-heavy, PK-heavy, in-over-his-head ref who always inserts yourself into the match somehow.  I said these things because I believed them, and I believed them because of the matches I have seen you work.  But I had no data to test this hunch... until now.  While I may not liked some of your calls, the idea in my head that you were a statistical outlier in those categories was untrue.

On Tuesday I found heaps of data that tracks referees' performances at: http://www.transfermarkt.co.uk.  Let me immediately and formally thank TransferMarkt UK (@TMuk_news); this article would not be possible without their free and easy to use site.  Armed with new info, I was able to do something that I've wanted to do for a long time- compare stats of MLS referees. (Nerd alert!) My ground rules:

• To ensure there was a valid sample size for any given ref, I needed to set a minimum number of matches worked.  34, an entire MLS season's worth, felt like a good number.  (Note: MLS refs don't work each week.  Even the busiest ones work only around 20 games/year).
• Only MLS regular season games would be in the analysis.  [Disclaimer: Use of the word "experience" below refers only to MLS games worked.  Other leagues/tournament matches that the MLS officials may have conducted were not factored into this study]
• I'd review cards issued and PKs awarded, all on a per match basis.
• All data is current as of 5/1/13

This left me with 10 MLS refs and 1021 games in the study.  I couldn't wait to see the results.

Yellows/Match - 3.3
The most benign of the stats to compare was yellow cards.  The overall average was 3.3 cards/match.  Among the refs, 3 of the 4 highest rates on this list were 3 of the 4 refs with the least MLS games under their belt.  Are yellows thrown around willy-nilly by newer refs?  Is it like being a parent, where you freak out at everything at first, but learn to go with the flow by the 3rd kid?  Maybe.  However, the least experienced ref in the study, Chris Penso, had the lowest of all ten with only 2.8/match.

 

Straight Reds/Match - .16
1st, some trivia - we learn that a straight red in MLS is 20 times less likely in a match than a yellow.  I never really thought about where that number would lie, but there it is.

As with yellow cards, experience seems to be a factor in how likely a ref is to issue a straight red (SR).  Is it such that, when considering the heavy consequences that come with a red card, the most experienced referees are more likely to have the confidence to send a player off?  The 3 least likely refs to issue a straight red were also the 3 with the least experience in MLS.  It's worthwhile to note that SRs are infrequent, but I still found it a little disconcerting that there would be so much of a difference between those most and least likely to issue one.

Total Reds/Match - .24 
Next, I made up a term called "Total Reds" (TRs).  TRs sums the red cards delivered both by 2nd yellows and by straight red cards.  I debated how to treat 2nd yellows, as technically they're supposed to be fouls of the same danger/caliber as any other yellow.  In real life, though, referees carefully deliberate before re-carding a player and sending him off.  Out of this study was born another new term, the "Naughty List".  I'll get to that in a minute, but first...

Leading the pack here is Baldomero Toledo.  In his matches, one way or another, someone gets tossed in nearly 1 of 3 matches.  Yes, there are 2 or even 3 red cards issues in some games, but for the most part only 1 red is ever given in a match.  At .32 TRs/match, that's about 7 guys a year that Toledo tosses.  In contrast, the relatively new Edvin Juresevic sees red in only 1 of 14 matches.  While I would normally be tempted to throw in disclaimer about how data can be interpreted many ways... wow.  That's just way different.

When looking at the relationship between yellows issued and 2nd yellows issued, I wondered if certain refs were more likely to card a player a second time than other refs.  As mechanical as you may try to act as an official, you're human after all.  Indeed, some refs are clearly more likely to show you red if you've already made it to the "Naughty List".

Use of Naughty List - 2.3%  In MLS, if you've already picked up a yellow card you've only got about a 1 in 44 chance of getting tossed.  Now, of course most players alter their behavior after being carded in order to avoid a second.  Also, coaching strategies may tilt towards pulling players holding cards off the field.  Still, there is the question: Once you have a yellow card, are you more likely to eventually get thrown out by certain officials more than others?  Even if unintentional, the data seems to indicate yes. 

Take the case of Chris Penso.  At .08 straight red cards/match, he has issued the 3rd least Straight Reds/Match in the study.  However, when it comes to Total Reds, he's in the middle of the pack.  That's because he issues a higher amount of 2nd yellow cards than anyone else.  In fact, it's not even close.  MLS players, be warned; don't get on Penso's bad side.  At the other end of the spectrum is "Cool Eddie" Juresevic, who barely ever issues a second yellow; he has done so only once in 194 cases.

PKs/Match - .22
Now let's address the most divisive issue a referee faces - awarding a penalty.  About 75% of awarded PKs are scored, so the decision couldn't be more critical to the outcome of a match.

Once again, experience comes into play.  Similar to issuing red cards, the more experienced referees were more likely to award PKs, while less experienced refs were not.  The 4 officials that awarded PKs most frequently were 4 of the 5 with the most games under their belts; only Kevin Stott did not appear in that group.  Does that mean that there is a crisis of confidence with those less experienced, or does it signal a move by MLS or PRO to reduce the overall occurrences, which would skew the stats downward for the newer folks?  Specifically, we can see Ricardo Salazar has reduced his rate each of the last 4 seasons, from a 2009 high of .5/Match to 2012's rate of only .12/Match.  Whatever the reasons are, the data tells us that the 3 refs who award the most PKs are 2.7 times more likely to do so than the 3 who are least likely.  Again, that strikes me as a significant difference.

So, is that normal?  And what is normal, anyway?  Is there a worldwide standard to which MLS refs could be compared? If there is one, I don't know where it exists.  So, I made it.

The Electrifying Conclusion - To see how MLS ref stats fared against those from leagues around the world, I picked 10 officials from 9 different leagues and performed the same analysis.  The first ref I chose was Howard Webb, who oversaw the 2010 World Cup final.  From there, I picked one more BPL referee, plus 8 more officials from various other leagues.  This resulted in an even bigger pool of games to review (1700+), which would only bolster my confidence in the figures.

Again, I started with yellow cards.  At 3.7 yellows/match, the World group looked more likely to dish out warnings than their MLS counterparts.  Interestingly, though, they were also less likely to issue straight reds.  The lesson I get out of that is this: More early cards controls violent behavior, which is responsible for most red cards.  Then something really, really freakin' cool happened...

As I finished the calculations for Total Reds/Match and PKs/Match, I saw both figures unite.  The TRs/Match were so close together that I pushed the decimal point out to the thousandths point, only to see that it was just barely far enough.  The difference in Total Reds/Match was 1 one thousandth of a point.  In the next column, the frequency of PKs/Match were also damn near identical; only 1 one hundredth of a point separated the two groups.  Over a total of 2731 matches analyzed, when comparing MLS referees to those around the world, there were no differences in frequency of players getting red carded during a match or penalties being awarded.

To be sure, officials should be measured mostly on the accuracy of their work, not on stats.
However, I do believe in the power of big numbers, and that lists such as these can identify outliers.
To me, this data drives home two messages.  One is that MLS does not have a group of referees that conducts matches wildly different than in other countries.  The other is that differences between officials are not due to anything other than the fact that they're people.  Some refs are going to be more likely to yellow card you, but they're probably not going to dish out any reds.  Others may never award PKs.  These differences may appear inconsistent, but just like a bumpy pitch, inconsistency is a part of the game.

So where Mr. Geiger may tend to give more yellows than the mean, and he may be 3rd most likely MLS ref on this list to award a PK, he was right down the middle on red cards.  Nothing I found indicates that he calls a match wildly differently than anyone else.

I cherish and appreciate referees for the service they provide to my soccer-playing country.  Much like coaches, they give of themselves so that current and future generations can enjoy the beautiful game.  When a referee is on the field, I play with a different urgency.  They make it real.  They are, in fact, a critical piece of the soccer landscape.  I should more often remember what my high school coach taught me to do after matches, which is to thank the refs.  Thanks, guys.
-@4castnate

The full tables, sorted by PKs/Match: