Category: Expected Goals

Gluck: Fourth Year Anniversary Edition

My thanks to everyone who has supported my web site the last four years!

It’s been a learning experience for me and, I hope, for you too.

As the new year starts I’ve got at least five new articles planned; here’s a quick synopsis on what to expect:

  • Following up on Coaching Youth Soccer Part I and Coaching Youth Soccer Part II, I’ll be offering Coaching Youth Soccer Part III – digging into which team statistics to use, why, when, and how to use them.  For those who don’t know me these three articles highlight my coaching philosophy into one three word catchphrase “muscle memory mentality“.
  • Two new individual soccer statistics:   This (may?) be controversial – My intent is to submit two new, professional level, individual, soccer statistics that could transform the player market value system.

Said differently; are private statistics companies, like Prozone Sports, OPTA, and InStat (along with player agents) manipulating the player market value system by ignoring what might be the most logical, intuitive, individual soccer statistics ever?

  • Expected Points – An updated version of my previously created Expected Wins series of articles.  A follow on to what was offered at the World Conference on Science & Soccer 2017, Rennes, France.
  • Expected Goals – A new way to calculate this over-hyped soccer statistic that brings it a bit closer to reality.
  • World Cup 2018 Total Soccer Index; to include predicting the winners after round one is complete.

For now, in case you missed one or two, here’s my rundown on the top five articles in each of the last four years.

In Closing:

  • I called for Jurgen Klinsmann to be sacked after WC 2014 because his tactics and in-game adjustments weren’t up to snuff.  Three years later the rest of the american mainstream soccer media world agreed and Klinsmann was sacked.
  • I called for Sunil Gulati to be ‘ousted’ after WC 2014 because his leadership in helping youth development and head coach selection weren’t up to snuff. Three years later the rest of the american mainstream soccer media world agreed and Gulati is out.
  • In hindsight – I wonder where we’d be in youth soccer development if we’d have made those decisions three years ago?
  • No, I do not favor Caleb Porter as the next US Men’s National Team head coach.  I like Caleb, he’s a stand-up guy and always took time to share and listen.  That said, in my opinion, he’s not (consistently) good enough at reading in game situations and making tactical adjustments that lead to better performances; the exact same issue I had with Jurgen Klinsmann.  .
  • I’m hopeful either Eric Wynalda or Steve Gans are elected as the next United States Soccer Federation President; electing Kathy Carter is a NO-GO in my view as there’s perceived ‘collusion’ between MLS and SUM.  As a retired Air-Force veteran perception is reality until proven otherwise – some may disagree?

I wish you all the best for the new year.

Best,

CoachChrisGluck

 

Gluck: Heaps of News in Major League Soccer This Week

Lots going on to share with you as Major League Soccer gets set for this weekend.  In particular order:

  • Major League Soccer Total Soccer Index (TSI)
    • Eastern Conference
    • Western Conference
  • Jay Heaps gets the heave-ho from New England; why?
  • Quality in MLS – has it got better since 2014?
    • If so, where?
  • Predictions for this weekend.
  • Closing thoughts on Expected Goals

As a reminder – I called out Expected Goals and Expected Passes this week.  Positive response from my European readers has been tremendous; so far my readers in the United States have remained quiet.

In case you missed it the explanation about what the Total Soccer Index is, is here.

Major League Soccer TSI:

This is how the league looks in a single table format; of course it’s pear-shaped from the start because the league doesn’t play a balanced schedule for everyone.

  • The hammer identifies teams who have sacked their Head Coaches this year; are Jim Curtin, Ben Oslen, and/or Jason Kreis on the block too?
  • The correlation of TSI to points earned is .82 this year; that’s an increase from the last two years.
  • Offering, in my view, parity is decreasing.
  • More to follow when we look at quality across MLS a bit later…

Eastern Conference:

The Eastern Conference has been the more predictable conference all year even though MLS has an unbalanced schedule.

  • Teams that usually possess the ball more, penetrate more, while showing greater patience in shot creation, end up with more goals scored.
  • This pattern, across all the categories in Possession with Purpose, more closely matches European League performances measured in the past.
  • Is this an indicator parts of Major League Soccer are growing closer to European Soccer in terms of tactics and how those general tactics drive similar results?
  • Maybe…
  • More to follow when looking at quality across the entire league.

Who finishes as Eastern Conference Champion?   Toronto.

Western Conference:

I’m not sure anything is settled in the wild-wild west.

  • We’ve seen musical chairs in almost every position of their conference table.
  • About the only thing remaining constant is the poor play of Colorado Rapids, Minnesota United, and LA Galaxy.
  • The greatest surprise may be the demise of FC Dallas, we’ve seen them swoon in late season before, does it happen again this year?
  • If any one team has been consistent this year it’s Sporting KC – but that’s the case every year.
  • With US Men’s National Team looking for a new Head Coach, after WC 2018, has Peter Vermes put himself in pole position over someone like Oscar Pareja?

Who finishes as Western Conference Champion?  Sporting KC

Who wins the MLS League Championship?  I have no idea.

Jay Heaps gets the heave-ho by New England, why?

Observations:

  • Their Attack:
    • 2nd worst percentage in overall possession across MLS
    • Mid-table in passing accuracy percentage
    • 3rd highest percentage in overall penetration of final third
    • 7th lowest percentage in shot creation
    • 5th highest percentage in shot precision
    • 8th lowest percentage in shot finishing
  • Opponents Have:
    • 2nd highest percentage of possession vs NER
    • Highest percentage of Passing Accuracy in MLS vs NER
    • Mid-table percentage in penetration vs NER
    • Lowest percentage of shot creation in MLS vs NER
    • Eighth highest percentage in shot precision vs NER
    • Fourth highest percentage in shot finishing vs NER

Conclusions:

  • Their Attack:
    • The team does not lack in attack.
    • Shot creation is at a lower level relative to a higher level of penetration; usually a positive sign of patience in attack.
    • That, coupled with being eighth highest in shot precision means when they create space there are putting shots on goal.
    • What (may?) lack in attack is finishing…. but when you look at the stable of players and see Kamara on 11 goals, Nguyen on nine, and Agudelo on eight they are pretty good/versatile in attack.
  • Their Defense:
    • Lacks by a considerable margin compared to their opponents.
    • Opponent’s are averaging over 80% passing accuracy; partly due to Revolution tactics of ceding space outside the final third in order to facilitate a better counter-attack.
    • What is striking is their opponents are also eighth best in putting shots on goal and fourth best in finishing.
    • That indicates Revolution opponents are gaining solid possession time BOTH INSIDE and OUTSIDE their defending final third.
  • Is it the wrong players on the pitch?
  • Is it the wrong defensive tactics on the pitch?

I’d say it’s both.

Quality in Major League Soccer:

It appears that quality has been roughly the same, year in and year out since 2014.

  • But that’s deceptive.  From 2014 through to 2017
    • The difference between average passing accuracy for the best and worst has increased from 7.17% to 9.50%
    • The difference between average penetration percentages for the best and worst has increased from 12.67% to 16.05%
    • The difference between average creation percentages for the best and worst has increased from 6.67% to 11.48%
    • The difference between average precision percentages for the best and worst has increased from 9.31% to 12.87%
    • The difference between average finishing percentages for the best and worst has increased from 21.73% to 23.42%
  • The gap between better teams and worse teams has widened.
  • Another indicator parity has decreased, not increased.

Given the trends offered through PWP analysis it appears parity is on the decline in MLS.

When the season ends poor management will be rewarded with more money instead of being relegated; entitlement is alive and strong in Major League Soccer.

Predictions for this weekend:

As with most weeks, the home team is more likely to earn points.

  • So far this year the home teams have earned 589 points versus 284 for away teams.
  • That’s a pretty solid 2 to 1 margin in favor of the home team.
  • Last year home teams earned 612 points compared to 300 points for away teams.
  • In 2015 it was 624 points for home teams and 324 points for away teams.
  • In 2014 it was 557 points for home teams and 323 points for away teams.
  • Conclusion – even without using Expected Goals it’s pretty clear a novice in soccer can guess who will earn points in MLS games.

By the way, if using the TSI to predict who would have won the U.S. Open Cup the numbers show Sporting KC with an average TSI of .41 (at home) while the New York Red Bulls were .00 (away from home).

The final result was 2-1 Sporting KC.  In hind sight the TSI predictor was accurate in predicting the U.S. Open Cup winner.

In closing:

  • Do you really need to know what Expected Goals are to predict which teams in Major League Soccer will earn points week to week?  No….
  • If you bet the home team you’ll be right roughly 66% of the time.
    • Just another reason to debunk the value of expected goals.
    • Oh… I’m hearing expected goals statistics are being used to predict results for the next year, using previous years data.
    • And that those correlations are pretty solid from year to year.
    • Well, they will be.
    • You’re only using one event-based statistic to predict results in the next year and that number is notoriously low for every team; for room for error is minimal.
    • I’m willing to bet a teams’ Expected Goals from two, three, or even four years ago will also have a pretty high correlation to the current year too…
    • Why?
    • Because only one variable is being measured and the variation in that variable is low – very low.
    • What makes that approach worse is it violates common sense.
    • Teams change players and Head Coaches from year to year and while they may score the same amount of goals, year in and year out, their overall results may be different because they got better defenders or improved their defensive tactics.
  • Parity in Major League Soccer has waned this year and it’s likely to get worse next year as LA adds another team.

Best, Chris

@CoachChrisGluck