Friday, May 28, 2010

Luchametric MLS Weekend Fixtures Handicapping: Crew, Houston, RSL, and San Jose

By MJD

Nordecke Luchador has evolved over the last year into its own unique brand.  From the heady daze of Gonzo Crew fanaticism, we now offer EPL and MLS statistical analysis, EPL, MLS and World Cup handicapping as well as tactical insight into the beautiful game.


In the spirit of Kaizen, the Japanese virtue of "continuous improvement," we offer our latest data chart for our loyal readers.

To the left are this weekend's fixtures in the MLS.  For each match we provide the moneyline and the decimal odds.  Next, we provide our readers with how these odds translate into probabilities.  For example, in the first fixture, D.C. is being given +175 or a 36% chance at a home win;  Chivas is being given +192 or a 34% chance of an away win.   IMPORTANT! Punters should always remember that these percentages should be compared to the historical probabilities that 50% of all professional matches end in a home win; 25% in an away win; and 25% in a draw--unless there is a substantial difference in team strengths.

So, for this fixture, the books have lowered the chance of a D.C. home win by 14 percentage points and raised the chance of a Chivas road win by 9 percentage points.  The chance for a draw is thus around 30%.

But we go even further.  The fourth column tells the prospective punter the percentage payout of their wager.  Thus, a successful bet on D.C. will earn you 175% of your initial bet; a successful bet on Chivas will earn you 192%.  So a $10 bet on D.C. gets you 17.50. Nice.

But wait! We then provide you with our own analysis of the value of the wager.  As a general rule of thumb, smart punters should compare the probability offered by the books to their own estimation of the probability of bet being successful.  Continuing with our example, we need to ask ourselves the following: "Ok, all things being equal, the home team should win 50% of the time. However, D.C. is a weaker team than Chivas.  That means that the chance of D.C. winning is less than 50%. The books say that it's as bad as 36%.  Do I think its worse or better than 36%?"

Answer?  If you think there is a 5% or higher difference between the probability offered by the books and your own independent estimation of the probability of a team winning, then that bet is most likely what the professional punters call "a value bet."  Here's why.

If the books say D.C. has only a 36% chance of winning, but you think their chance is actually 41% or higher, then the amount of money the books are offering you is actually higher than it "should" be.  Remember, the higher the probability of a win, the lower the books pay out.

Further, a value bet is one where the amount of risk is "worth" the potential payout.  Bookies want to offer you less money for your risk than is "fair" or "real."  So, if you can find a match where you believe the books have it wrong and the actual probability of your team winning is higher than they've determined, then you've found a bet that will pay out more for your risk than the books would pay out if they had gotten the odds right in the first place.  In D.C.'s case, the books have set the odds at 36% and are offering you a payout of 175%.  If the books thought the probability was 41% or greater, they would pay winners less than 175%  As such, if you've decided that D.C. actually has a 41% chance or higher (let's face it, Chivas isn't very good) then taking D.C. to win at home would be a value bet. 

But what does Nordecke Luchador think about this match?

Our final column provides you with our advice for every match this weekend.  Our recommendations are based on A) our determination of the value of the odds being offered; and B) our usual statistical comparison of the two clubs including the Luchametric power rankings.  If we recommend a team, we think its a value bet with a high probability relative to the risk.  This week, we think C-Bus, Houston, RSL and San Jose all fit the bill.  And we think the D.C. v Chivas match should be avoided like the plague.

And remember, we don't recommend parlays ever.  Do your homework and pick your spots.  For example, Donovan is up with the USMNT.  We think that makes the Crew an even more valuable bet against LA than the numbers alone might suggest.  However, Houston is being given a 68% chance of a win, and we think it's definitely even higher. Indeed, this match probably represents the bet with the highest probability and the lowest risk of the weekend.  But the probability of both Houston and the Crew winning in a parlay is a rather low 30%.  If you bump the Crew's probability up to 51%, the parlay would still only be at 35%.  The probability of Houston, the Crew AND San Jose all winning? 17%.  RSL actually represents the wager with the highest value (in our view) as the payout is 69% on your wager (compared to 46% for Houston) and we think the probability of a RSL win could be as much as 10 percentage points higher than what the books have it at.  That's some sweet action.

Finally, it's important to note that we've excluded draws in our chart because draws never, ever offer high percentage probabilities nor value bets.  That's a maxim of betting on soccer we've learned from our secondary research (seriously).  And you can take that to the bank. C-Bus 'till I die!

Thursday, May 27, 2010

Luchametric USMNT Tactical Analysis: What to Look for Against Turkey

By MJD
Alright, so now we know which Yanks will be heading to South Africa, but what should we expect to see tactically, hopefully starting Saturday against Turkey?

We've looked through the lineups at football-lineups.com for both the 2009 Confederations Cup and the 2010 World Cup Qualifiers, and what follows is our tentative projection of what to expect from Bradley in the two upcoming friendlies and against England on June 12th.  We went with the above two sets of matches as we think the Gold Cup in 2009 and the various friendlies over the last year don't accurately represent our best XI (in fact, we think our B-team was a grave disappointment in virtually every match it played).

The first graphic (above) is our prediction if Bradley runs out a standard 4-4-2.  In such games, he has historically put Dempsey and Donovan on the right and left flanks respectively.  Clark and Bradley are the two most common holding/defensive mids in Bradley's various lineups.  We think Altidore is a sure starter, and we're going with Gomez up front with him. Why? We think the combination of Altidore's physicality and ability to hold the ball combined with Gomez's speed and technique will give the US two very different looks at striker--which we dig like dirty rice. Note that Goodson could start for Onyewu and Bocanegra for Bornstein. In our review of Bradley's lineups, the back four were very frequently changed up and moved around. We don't think that will change.  If Gooch isn't ready by June 12th, you could very well see Bornstein, Bocanegra, DeMerit and Spector filling out the back four.

Next, Bradley has also used a rather uncommon 4-2-2-2.  In the graphic to the right, you'll see that in this schema, Bradley and Clark drop back in front of the back four, and Donovan and Dempsey switch sides.  Donovan played mostly on the left with Everton this winter and spring, and this may affect Bradley's tactical decisions.  Regardless, we've still got Gomez and Altidore up top.  At Puebla, Gomez was typically the lone striker, and for the US Altidore has played on the both the right and the left, so it will be interesting to see how Bradley decides to use the two forwards (if he goes with Gomez and not Buddle or Findlay).

Finally, we came up with our own lineup, featuring the return of the prodigal son, DaMarcus Beasley.  He's been in our doghouse since the last World Cup, and he's been in Bradley's for about a year, but he looked very, very good against the Czechs.  The last graphic is our hypothetical XI with Beasley in the mix--in this case, a 4-4-1-1.


In this formation, we put Beasley on the right wing, moving Dempsey over next to Donovan with Clark as the sole defensive mid.  Up top we push Altidore ahead of Gomez, allowing Gomez space to create.  We think this is a great attacking formation for the Yanks.  We look forward to seeing what Bradley's actually going to do with Beasley.  For example, Bease could replace Clark or Bradley in the 4-4-2 or the 4-2-2-2.  Also, we haven't even considered the option of pushing Dempsey or Donovan up top as one or both strikers, which would give the US another and very different look.

Needless to say, it's nice to see that the US has some very tantalizing options.  Word.

Tuesday, May 25, 2010

MLS Power Rankings Updated: The Big Two

By MJD

I've been out of the office for the last week but have finally had a chance to catch up on things in the MLS. We suggested in an earlier post that MLS parity might be a thing of the past, and the updated numbers continue to support that claim.

Both LA and the Crew enjoy a substantial advantage in the LM and especially in their respective Z-scores relative to the rest of the league.  Note also that there seems to be a large number of middling teams and three to five absolute bottom feeders.  Here's how we'd divide up the league:

The Big Two: LA and C-Bus
The Great Pretenders: RSL, SJO.
The Mighty Middle: Dallas, Colorado, Houston, Chicago, Toronto, New York, Seattle, Chivas, KC
The Vanquished if Not Relegated: Philly and D.C.

You can slice and dice it however you'd like, but there clearly has been a shift in the relative strengths of the different clubs.  Finally, the West continues to dominate the rankings and the single table.  Six out of the eight top teams in the LM are from the West; as are five out of the top eight in the single table.  Again, it ain't about the two conferences, which are a joke, it just adds to the overall impression that the era of parity in MLS is over.

Saturday, May 15, 2010

Crew vs Chivas USA Statistical Breakdown

By MJD

Yesterday we gave you a breakdown of the Crew's tactics, today we give you a quick statistical comparison of the two clubs.

This game is not a sure 3 PTS for C-Bus, but it is 3 PTS the Crew certainly should earn.   Last year Chivas USA had only 6 draws, lowest among the 8 playoff teams.  This year they have 3 wins and 4 losses  and only 1 draw.  But they've scored 5 goals at home and 5 goals on the road.  In short, Chivas is a club that so far has been inconsistent--but they are dangerous depending on which club shows up for that particular match.

Notice that Chivas allows fewer SHTAs and SOGAs per game than does the Crew. As we said earlier in the season, the Crew's defense gives us cause for concern, and last week did nothing to alleviate those anxieties. However, the rest of the numbers tell us that the Crew enjoy a substantial advantage including GF/GM, GA/GM and PTS/GM. Most importantly, in the LM Power Rankings, the Crew enjoys a considerable advantage of over 1.5 STDDEVs.

Here's hoping that Hesmer has his head screwed on straight, that the Crew's attack gets more efficient (several players last week failed to finish exceptionally good chances) and that during the defensive phase the Crew begins to show signs of being able to completely shut down lesser teams . Columbus 'til I die.

Friday, May 14, 2010

Crew Tactical Breakdown: What to Look for Against Chivas

By MJD

The Crew will most likely line  up in their standard 4-4-1-1 on Saturday night, with Rogers and Gaven on the wings, and Padula and Hejduk as the outside backs.  Iro, Marshall or O'Rourke will start as one of the center backs opposite Brunner.  And look for Carroll and Moffatt to play as the holding center-mids.

In this formation, GBS plays rather high, which has been the subject of some discussion in the Crew blogosphere.  The argument is that Guillermo is better utilized as a more traditional number 10, roaming the center which allows him to create more.  In such a formation, the Crew would most likely go to a 4-3-3 or a 4-4-2-- in the latter they'd need to go with two forwards or a forward and striker combo: Lenhart and Ekpo, for example, or Lenhart and Gaven, or Lenhart and Garey.

However, as the next graphic shows, even in RW's present schema, Schelotto is still being used as a creator.  Most importantly, the Crew's attack is a little more complicated and (we think) more dynamic and thus more difficult to defend.

In the graphic to the right, you see the Crew in the attack phase. The graphic represents what the club was trying to do last Saturday against New England and can be considered RW's basic approach on attack, at least with these players.

Notice that in this particular version of the Crew's 4-4-1-1, either Padula or Hejduk pushes up, in this case we'll push up Padula.  But the left back is not merely crashing down the wing.  Rather, in this scenario, the Crew is really playing a 3-2-3-2 or some might call it a 3-2-4-1. Regardless, the system is designed to get the left or right back to work with the holding mids, the wing AND the number 10, creating multiple passing possibilities and thus maintaining possession.  All the while, they are building up the attack gradually, looking to get Rogers and Lenhart to angle in on the box and finish, with both Padula and Schelotto providing service.

As you can see, the interaction among the back, the mids, the "10" and the wing creates multiple and dynamic passing combinations, and can easily be shifted to the opposite wing with the right back (Hejduk) moving forward instead of the left.

Note also how this approach allows the team to get 5 players forward and that GBS is still in a central position.  Further, if a run takes Rogers over to the right, Gaven and GBS can rotate left.  Finally, Carroll and Moffatt are in a position to both support the attack and to drop back in front of the back three if the Crew lose possession.

Three things to take away from last week's game in light of the above analysis.

First, the striker absolutely must finish efficiently for the formation to produce goals.  Lenhart did not do so last week in spite of some excellent service from both GBS and the wings.  Overall, Lenhart has converted 40% of his SHTS to SOGs, but has scored only 1 goal in 4 SOG total or 25%.  That's average for the league overall and below average for forwards.  Perhaps last week was on off night.  But maybe RW should start Ekpo up top instead of the big blond..

Second, the Crew looked vulnerable on the counter-attack last week, and so the two central mids have to be fast enough to respond quickly when the team is forced to defend.

Finally, this approach leaves the Crew with only a three-man back line.  If the team can delay up high and thus allow players to get back and defend as a unit, this is not so dangerous.  But make no mistake, this is an attacking formation that relies on possession and a quick retreat by the mids and the attacking back(s)  to insure the opposing team doesn't overwhelm the three defenders.

Chivas has played a 4-2-3-1 its last two matches, and RWs back line will need to be sharp, even at home, if the Crew hopes to avoid another see-saw match against a lesser club. It will be interesting to see if the combination of Chivas' two mids in the 4-2-3-1 and the Crew's attacking 3-2-3-2 creates space in the middle of the field and thus a lot of movement and wide open play.

Our prediction? Look for the Crew to continue to gel as they get more games under their belt and for Chivas to give up 2 goals at least.  Note that historically Chivas does not draw very often.  Final score 2-0 Crew.  Massive. Hagado C-Bus.

Tuesday, May 11, 2010

MLS Power Rankings Updated: Go West My Tan-Faced Children

By MJD
We loathe the International House of Soccer's insistence on the two divisions as it distorts the league standings and has virtually no impact whatsoever on who gets into the playoffs.  And although relegation is a long way off (if it ever happens in our lifetime--hopefully another 40 years or so), we hope the league will go to a single table much sooner.  But this post isn't really about any of that.  It's about MLS on television and the land of great Mexican food, medical marijuana, and the EPL "late" game with breakfast.

Take a look at our Power Rankings and the Single Table and you'll notice that the western teams currently dominate the league.  In the Power Rankings, 5 of the top 8 teams are out of the West; and in the single table 6 out of the top 8 teams play their home games west of the Mississippi.  Again, we have disdain for MLS' marketing strategy, but we think it's worth noticing that the best football in the league is currently being played far away from the Northeast corridor.  Other than the Red Bulls, the top teams in the "East" are actually in the Midwest.  And the majority of top teams in the league are either in the same time zone as Chicago or even further away from the nation's most densely populated region. Of course, the big standout exception is our beloved Crew who continue to contend for a third-straight Supporters Shield, which would do more than anything to demonstrate the invalidity of the two-division playoff scheme. (That's a post for another day).

MLS' business model is designed to insure big market teams like DC, NY, New England (Boston), Philly and Toronto can carry the load when it comes to television market share.  Let's face it, the Crew vs. New England on a Saturday night is one thing; the Crew vs. KC is something else.  The former includes greater Boston as one half of its potential audience, which is the nation's 7th largest television market; the latter includes the nation's 34th and 32nd largest markets respectively (who knew KC was bigger than C-Bus?).

All of this is simply to say that you can bet the suits at the IHOS are sweating a bit as the season unfolds.  Philly looks like they genuinely stink.  DC is drawing fewer of its rabid white-collar suburbanites out for a night of pseudo-hooliganism than their reputation would predict, and NE looks like its headed for yet another season of marginality.

As such, you might think we'd see more games on FSC from the major markets in the West this year than we would in the East.  Unfortunately, it's more likely Don Garber and the Sports Admin majors in the league offices will decide that mediocre football that caters to the fans in the Northeast corridor is a potentially more profitable product than great football that caters to the fan of the beautiful game.

Which brings us once more to the patron saint of American soccer: Papa Walt

Come my tan-faced children,

Follow well in order, get your weapons ready, Have you your pistols? have you your sharp-edged axes? Pioneers! O pioneers!

For we cannot tarry here, We must march my darlings, we must bear the brunt of danger, We the youthful sinewy races, all the rest on us depend, Pioneers! O pioneers!

O you youths, Western youths, So impatient, full of action, full of manly pride and friendship, Plain I see you Western youths, see you tramping with the foremost, Pioneers! O pioneers!

Saturday, May 8, 2010

Luchametric Power Rankings Updated: The MLS Big Three and the End of Parity

By MJD

It's early, it's early, um, AND it's still early. But the latest update of the Luchametric tells us that LA, the Crew, and the revived Red Bulls sit atop the league and have opened a substantial gap between the top three and the rest of the league.

The top three are all within .65 of a STD DEV of each other in the Z-Score and are all above 1500 in the actual LM score.  The rest of the league is below 1500, with Chicago closest at 1199, and only the top three are over 1.00 STD DEVs above the league average.

One thing we like about the LM is that it takes into account the different factors which distort the league point standings (most importantly, Games Played) and provides a single score that allows for a comparison of all the teams in the league that is more representative of a team's success than mere points or even Points per Game can provide.

We'd also point out that Chicago and San Jose have moved up in the rankings and should be treated with seriousness by team managers and punters alike.

Finally, the bottom three (Toronto, Philly, and DC) remain at the bottom and still are playing substantially poorer than the rest of the league.  There is much parity in the middle, but at this early stage, there is less parity across the entire league than we've seen in recent years.

Tuesday, May 4, 2010

Crew Shots Against Grid Updated

By MJD
We've updated our chart tracking the Crew's SHTA and SOGA. The boys are still giving up more SHTA and SOGA than the league average (13.25 vs 10.77 and 4.5 vs 4.37 respectively), and their actual GA compared to their Expected Goals Against remains good (3 GA vs 5 EGA).  But as the chart indicates, they're still allowing opponents to create quality chances against them--look at all those chances right inside the box--14 total right in front of the net.

The question, then, is two-fold, is the Crew's strategy to bend and not break and then counter-attack?  Or does the team need to address its defensive issues quickly, as they cannot hope to continue to allow this many chances per game over the course of the remainder of the season and not start giving up goals?  What if Hesmer goes down and his form drops a few notches?  Our hunch right now is that RW needs to look hard at his back line.  One solution would be to hold Hejduk back, for example, at least from run after deep run up the wing.  A few judicious runs forward? Fine.  But using him as a real attacking back?  I don't know.

Finally, we've updated the power rankings and made them simpler.
The LM is now a single score normed so that the worst team in the league is now a zero.  We hope it makes more sense.  And in spite of our misgivings about the Crew, they are still second in our rankings, which is a measure of production per game, not simply overall production.  As such, they are still right up there at the top of the league.