Fans' View

Fans’ View: We’re going streaking!

Can you believe the streak we were on at the beginning of the season? We hadn’t won a game “since August of 2016!”

How about the one we’re on now? “It’s the longest winning streak in Union history!”

We were horrible then. We’re unstoppable now, right? Not quite.

MLS is a league built on the principle of parity – any team can beat any other team on any given Sunday (or Saturday afternoon…or Wednesday night, or whenever MLS plays). Each team has an equal chance of winning; basically a coin flip. But if it really was 50/50, we wouldn’t see these storylines, right? I’m not so sure.

So I decided to do a little experiment of my own. Flip a coin 34 times and repeat that seven-fold, thus replicating the Union’s existence. The rules were simple: heads=win, tails=loss. If the season of 34 flips ended above .500, we make the playoffs; below .500 and we’re out.

The first season I didn’t know what to expect, and the results seemed to be fairly randomized, though there was a single streak of four wins in a row. Interesting, given that it matches the longest win streak in Union history. Despite that, the season ended 18-16, respectable for an expansion side new to the coin flipping game and pleasantly above the red line and into the playoffs.

The second season started woefully. The team lost 8 out of the first 9 matches, and I knew we wouldn’t make the playoffs. I started questioning my coin-flipping skills. Perhaps I wasn’t up to the task. But I stayed on track and regained a little respect with a 15-19 record. Sadly, no playoffs.

I decided I needed to try harder. From then on, rather than just randomly flipping, I would purposefully try to flip a heads for the win. The result? The third and fourth seasons ended with the same 15-19 losing records and again no playoffs.

The team needed a new approach, so I tried flipping with my left hand to start the 5th season. Five loses in a row later and I was resigned to being a losing coin-flipper. Then out of nowhere, a delightful stretch winning 9 of 10 had me thinking I had figured it out. We ended the season 17-17, but according to my rules, not being above .500, again missed out on the playoffs. Still, it seemed like progress.

My fan base (two elementary school sons) was getting restless. It was time for real change. I got a nice new shiny quarter, trimmed my thumbnail to a perfect curve, stood up, raised the window shades and really focused on flipping a heads. I made a critical tactical adjustment and decided to always start the coin heads up. To my amazement, we started the season… a perfect 10-0! I continued some quality tosses to end that 6th season an impressive 24-10, and for the first time since the inaugural season, back into the playoffs. We were going somewhere.

For the 7th and final season, I would follow the same plan, certain that I would continue to improve upon my developing mastery of the coin flip. We won the first 5 of 7 before a dreadful midseason run losing 8 out of 10. Perhaps we were a bit injury ridden (my thumb now showing fatigue). The season finished 16-18; out of the playoffs once again.

A team doomed to sub-par mediocrity and a coin-flipper that just can’t figure it out.

So what’s the point of all this? Given enough time, when observing a random process, you will see patterns that are unexpected to occur. There is a lot of inherit randomness in soccer to begin with. Many world football researchers estimate about 50% of the results in the top leagues are purely due to chance. Under the model of MLS, randomness accounts for even more of what we see.

Be wary of talk about streaks. It’s all about framing. When writing my account of the coin-flips, I looked for some outliers and I tried to make sense of them. That’s what our human brains are designed to do; see patterns out of randomness. When I said that the team lost 8 of 10, I’m purposefully ignoring the 5 of 7 wins before that. We all do the same thing with the Union. Especially when we add in draws, now we can say things like “haven’t lost” or “haven’t won” for much longer strings of results. We like to create storylines to make it more interesting, and perhaps some of those are true, but perhaps a lot is just due to chance.

What’s funny is that I actually thought I had control over it for a little while. I mean, that stretch of 10 heads-up in a row was something else, right? But it’s not some mastery of the flip – it’s just a rare, but normal variation. Ask a mathematician about the Poisson distribution if you don’t believe me.  DC United went from worst to first in the course of a season with the same coach. We all proclaimed Atlanta under Tata Martino was amazing; now they’re below the Union. Perhaps my thumb’s influence is equal to that of a coach in MLS – more perceived than actual.

For all the drama of my coin-flip experiment, guess what the final tally was over the seven seasons combined? 120-118 for a win percentage of 50.5%.

Streaks happen when you flip a coin flip. Enjoy the random roller coaster!

7 Comments

  1. Matt Thornton says:

    Nice one, Scott!

  2. Buccistick says:

    Ha ha! Thank you for this perspective, Scott — right down to the “fan base” 🙂
    .
    Reminds me of the time I played the chaos game/drew a Sierpinski gasket by hand.
    *
    … come to think of it, that might better model the three outcomes of win, lose, or draw that soccer produces.
    .
    Also, relevant: http://mathforum.org/library/drmath/view/72186.html

  3. el Pachyderm says:

    Funny stuff.
    .
    …and naturally there is no way to secure one point which probably accounts for some of those Tails flips accumulating crucial crucial points along the way.

  4. You’re leaving out a couple of key things in your coin flip analogy:

    1. MLS games are Home or Away. Neither is a 50-50 coin flip. Some of the “streakiness” relates to scheduling.

    2. Coin flips are completely independent of each other. Soccer games are not. Players learn, they gain or lose confidence, etc. Some of the streakiness is legitimate change.

  5. You should use a die and then make 1-2 a win, 3-4 a loss 5 a tie and then 6 a win if you’re home, and a loss if you’re away

Leave a Reply to AJ Cancel

Your email address will not be published. Required fields are marked *

*