2016 Playoff Projections, or Hope By The Numbers

[Gotham Gator]

Yes. This makes absolute sense. But.
I will always try to be the guy -- especially in sports -- who says streaks are random, and not due to coaching, new players, or clutch performance.
I could write pages on this but I'll just state my premises and conclusions: As a species we excel at noticing patterns and coming up with explanations for them. It's how we survived, prospered and developed. Smart people are good at it and dumb people are good at it. The explanations of smart people tend to make more sense and be right more often, but every one of us including the idiots can notice patterns and come up with plausible-sounding theories in our sleep. Otherwise we would lack both high technology and popular conspiracy theories. It's in our nature.
But any statistician on the planet will tell you that in any true random distribution you will find streaks, clusters and patterns that have no meaning or rationale whatsoever. And as a species we suck at admitting this in practical applications. We especially suck at it in sports, because we want our heroes to be clutch and our goats to have flaws. But much, much more often than we are willing to admit, the hot streaks, slumps, and performance in big moments just reflect random distributions.

This is completely true, but I do think there are good reasons to think we are a much better team now than we were when we only won 1 of our first 8 games.

I remember when I was in finance class in B School. The professor had everyone call out, in order, the result of an imaginary coin flip and wrote the sequence on the board. He then handed everyone (about 80 people) a nickel and had them flip it then had them call out the result - again, writing the sequence on the board. The main difference in the two sequences is that the result of the real coin flips had much longer streaks of heads or tails than the imaginary ones did. The humans in the room weren't comfortable with an objective 50-50 chance leading to long streaks with one result or the other. The coins themselves were fine with it.

 

[mgarbowski]

This is completely true, but I do think there are good reasons to think we are a much better team now than we were when we only won 1 of our first 8 games.

I remember when I was in finance class in B School. The professor had everyone call out, in order, the result of an imaginary coin flip and wrote the sequence on the board. He then handed everyone (about 80 people) a nickel and had them flip it then had them call out the result - again, writing the sequence on the board. The main difference in the two sequences is that the result of the real coin flips had much longer streaks of heads or tails than the imaginary ones did. The humans in the room weren't comfortable with an objective 50-50 chance leading to long streaks with one result or the other. The coins themselves were fine with it.

It's very possible you and Ulrich might be right in this instance. Just because we underestimate the effect of randomness doesn't mean it explains everything. Most of the time patterns have a reason which is why our tendency to look for and explain them works. But I'm still going to be the guy who reminds everyone that randomness explains more than we think.

 

[413Blue]

Streakiness can also be explained by familiarization by opponents. When we are winning, opponents begin to get plenty of film to develop a gameplan to beat us. Once they have figured us out, we then go on a losing streak until we find something that works, and then we go on a winning streak, and then they figure us out and develop a gamplan to beat, we then go on a losing streak until......
That is the key to this league, being able to adapt your system game to game to best defeat your opponents, it makes it difficult for your opponents to react to, and generally smoothes your overall results. This partially explains why the Galaxy have been so successful for so long is that they often build their lineup and gameplan each week instead simply trying to "play their game" each week. This also explains why the Red Bulls have had very little sustained traction since they came out of the gate on fire last season with the high press, it's been figured out now.

To the live action. The Union are currently up on DCU 3-0. This will put them 1 point behind us on even games played, but seven goals clear in GD. Montreal is away to RSL in about an hour. I'm planning to catch that match on MLS live! A win for Montreal would put them on top of the east by a tiny fraction in the PPG column, and at least a five goal advantage in the GD column.
The East is certainly within our grasp, but still a great challenge. The Supporter's shield would be an extreme mountain to climb as we would have to continue a solid run pretty much all season and hope to outpace several quality sides.

How are the CCL league spots determined again? MLS Cup, Supporters Shield, and runner-up correct? If the Western Conference team wins the Supporter's Shield and MLS Cup, then Eastern Conference champion would get the spot correct?

 

[adam]

Streakiness can also be explained by familiarization by opponents. When we are winning, opponents begin to get plenty of film to develop a gameplan to beat us. Once they have figured us out, we then go on a losing streak until we find something that works, and then we go on a winning streak, and then they figure us out and develop a gamplan to beat, we then go on a losing streak until......
That is the key to this league, being able to adapt your system game to game to best defeat your opponents, it makes it difficult for your opponents to react to, and generally smoothes your overall results. This partially explains why the Galaxy have been so successful for so long is that they often build their lineup and gameplan each week instead simply trying to "play their game" each week. This also explains why the Red Bulls have had very little sustained traction since they came out of the gate on fire last season with the high press, it's been figured out now.

To the live action. The Union are currently up on DCU 3-0. This will put them 1 point behind us on even games played, but seven goals clear in GD. Montreal is away to RSL in about an hour. I'm planning to catch that match on MLS live! A win for Montreal would put them on top of the east by a tiny fraction in the PPG column, and at least a five goal advantage in the GD column.
The East is certainly within our grasp, but still a great challenge. The Supporter's shield would be an extreme mountain to climb as we would have to continue a solid run pretty much all season and hope to outpace several quality sides.

How are the CCL league spots determined again? MLS Cup, Supporters Shield, and runner-up correct? If the Western Conference team wins the Supporter's Shield and MLS Cup, then Eastern Conference champion would get the spot correct?

Each conf winner, USOC, MLS cup. If any team wins 2 spots. The extra spot goes to next highest place is SS race who didn't qualify yet.

http://www.mlssoccer.com/post/2015/10/15/how-mls-clubs-qualify-2016-17-concacaf-champions-league

 

[mgarbowski]

Each conf winner, USOC, MLS cup. If any team wins 2 spots. The extra spot goes to next highest place is SS race who didn't qualify yet.

http://www.mlssoccer.com/post/2015/10/15/how-mls-clubs-qualify-2016-17-concacaf-champions-league

To be clear, conf winner means regular season not playoffs.

 

[adam]

To be clear, conf winner means regular season not playoffs.

Yes, regular season conference winner. And if a Canadian team wins either conference or MLS Cup, then it moves to the next in line for SS method.

Theoretically, Toronto could win the East, Vancouver the West and Montreal the MLS Cup. The 3 US teams below the SS winner (Toronto or Vancouver) would all go to the CCL. It's so contorted, it just reaks of MLS. Why can't Garber just do something the simple way?

 

[mgarbowski]

Yes, regular season conference winner. And if a Canadian team wins either conference or MLS Cup, then it moves to the next in line for SS method.

Theoretically, Toronto could win the East, Vancouver the West and Montreal the MLS Cup. The 3 US teams below the SS winner (Toronto or Vancouver) would all go to the CCL. It's so contorted, it just reaks of MLS. Why can't Garber just do something the simple way?

Try reading the rules about UEFA CL, Europa, the FA Cup, League Cup, & Prem League.

 

[Gotham Gator]

Worth noting that the first tiebreaker in MLS is number of wins, with goal differential coming afterward.

 

[mgarbowski]

Very busy so this will be short this week.
There was so much whining in the shoutbox last night. It was unbecoming. We had 1/3 possession, and were outshot almost 5-1. Yet apparently the only reason we lost was, (1) the ref, (2) the announcer, and (3) someone's least favorite player. We had a clearly tired team from 90º heat, 3 games in one week, and 10 hours in an airport. It happens.

We slipped to second place in PPG, but still look to be in solid shape overall. Most of the East will pick up a game on us this week as we are off midweek but Chicago, Columbus, Montreal, Orlando, RB, Toronto all play Wednesday. Come Sunday, Montreal should have the tired legs and our boys will be well rested.


After one win and one loss here's an idea of what the team needs to do to reach assorted levels.


Our West record drops to 3-2-0 (1.8 PPG) and Road sits at 5-3-1 (1.78). In the biggest game for us on Wednesday, let's go Portland.

 

[mgarbowski]

We're still in first place, and also a clear first via PPG.
We're up to 3rd place Supporters Shield, and tied for 4th by PPG. Those are new highs for the team.
Now up to 2 games over .500 against the East, and 1.50 PPG.
This is heady stuff. We were crap last year. Last year at this point we had 24 points, six wins, and Lampard hadn't shown up yet after 21 games.
Starting with the win against Vancouver, which ended the 7-game winless streak, Las Palomas have earned 2.0 PPG over 13 games. Keep that up over the last 13 games (that's a big IF) and that's 59 points.
On the road the Robber Barons are at 1.90 PPG for the whole season. That is better than the combined H/A PPG for every team in the league except Colorado. Three more road wins in 7 tries and we tie the "modern" MLS record. We have twice as many road wins as home wins. We have more road wins that 11 teams have home wins (12 if you count ourselves).
With 13 games to go, we need just 4 or maybe even just 3 more wins probably to get the Celestes a playoff spot.
We're aiming higher at this point.
After the next game against the Red Bulls there will no longer be a H/A imbalance in games played. That's bad/good news if you can figure it out.
We have a 3-point lead on the Union, and the closest thing to bad news is that they can erase that by winning their extra game remaining. The good news is that extra game is on the road. They have 6H left and 8A, and they are terrible on the road, with only one win in 9 tries so far.
The playoff line projection is trending down. Will it pick up again like last year? If it stays where it is we're virtually a playoff lock.


The what-we-need-to-do part is starting to look easy:

 

[LionNYC]

We're still in first place, and also a clear first via PPG.
We're up to 3rd place Supporters Shield, and tied for 4th by PPG. Those are new highs for the team.
Now up to 2 games over .500 against the East, and 1.50 PPG.
This is heady stuff. We were crap last year. Last year at this point we had 24 points, six wins, and Lampard hadn't shown up yet after 21 games.
Starting with the win against Vancouver, which ended the 7-game winless streak, Las Palomas have earned 2.0 PPG over 13 games. Keep that up over the last 13 games (that's a big IF) and that's 59 points.
On the road the Robber Barons are at 1.90 PPG for the whole season. That is better than the combined H/A PPG for every team in the league except Colorado. Three more road wins in 7 tries and we tie the "modern" MLS record. We have twice as many road wins as home wins. We have more road wins that 11 teams have home wins (12 if you count ourselves).
With 13 games to go, we need just 4 or maybe even just 3 more wins probably to get the Celestes a playoff spot.
We're aiming higher at this point.
After the next game against the Red Bulls there will no longer be a H/A imbalance in games played. That's bad/good news if you can figure it out.
We have a 3-point lead on the Union, and the closest thing to bad news is that they can erase that by winning their extra game remaining. The good news is that extra game is on the road. They have 6H left and 8A, and they are terrible on the road, with only one win in 9 tries so far.
The playoff line projection is trending down. Will it pick up again like last year? If it stays where it is we're virtually a playoff lock.
View attachment 5331

The what-we-need-to-do part is starting to look easy:
View attachment 5332

As these posts continue as the weeks go by, I'm expecting new nicknames.

 

[CCMore]

We're still in first place, and also a clear first via PPG.
We're up to 3rd place Supporters Shield, and tied for 4th by PPG. Those are new highs for the team.
Now up to 2 games over .500 against the East, and 1.50 PPG.
This is heady stuff. We were crap last year. Last year at this point we had 24 points, six wins, and Lampard hadn't shown up yet after 21 games.
Starting with the win against Vancouver, which ended the 7-game winless streak, Las Palomas have earned 2.0 PPG over 13 games. Keep that up over the last 13 games (that's a big IF) and that's 59 points.
On the road the Robber Barons are at 1.90 PPG for the whole season. That is better than the combined H/A PPG for every team in the league except Colorado. Three more road wins in 7 tries and we tie the "modern" MLS record. We have twice as many road wins as home wins. We have more road wins that 11 teams have home wins (12 if you count ourselves).
With 13 games to go, we need just 4 or maybe even just 3 more wins probably to get the Celestes a playoff spot.
We're aiming higher at this point.
After the next game against the Red Bulls there will no longer be a H/A imbalance in games played. That's bad/good news if you can figure it out.
We have a 3-point lead on the Union, and the closest thing to bad news is that they can erase that by winning their extra game remaining. The good news is that extra game is on the road. They have 6H left and 8A, and they are terrible on the road, with only one win in 9 tries so far.
The playoff line projection is trending down. Will it pick up again like last year? If it stays where it is we're virtually a playoff lock.
View attachment 5331

The what-we-need-to-do part is starting to look easy:
View attachment 5332

Just posing a question. for those mathematicians out there who for some reason enjoy probability equations. What is the probability we make the play offs this year?

 

[Seth]

As these posts continue as the weeks go by, I'm expecting new nicknames.

Even with nicknames we already know, I was pretty impressed there was a different one in each paragraph.

 

[mgarbowski]

Just posing a question. for those mathematicians out there who for some reason enjoy probability equations. What is the probability we make the play offs this year?

There are a couple of sites that do that and are both pretty good. Midas Mulligan found both last year. The first is
http://playoffstatus.com/mls/easternstandings.html
It is already updated through last night and gives us 94% of making the playoffs, and 54% of finishing first or second.
The second site is http://www.sportsclubstats.com/USA/NYCFC.html. It has lots of fun ways to break things down, including a much more detailed playoff line breakdown. It's problem is (1) it does not update regularly and sometimes not for a few weeks, and (2) even when it is on schedule it does the week on an annoying Sunday thru Saturday basis even though every MLS week ends on Sunday. It is currently now up-to-date through Friday. For us that means it has the SKC game but not yesterday and has the playoff chances at 88.8% which is really in line with the other site. But it also shows, for example, that if we finish with 55 points there is a 61% chance we finish first in the East and 1% chance of Supporters Shield. At 59 points there would be a 24.2% chance we win SS. One of the most fun features here is showing the effect that the result of upcoming games will have on playoff chances. But because it is behind schedule that's not as much fun. For example, it says a loss to the Impact would knock the playoff chances down 2.9% and a win raises it 7.2%.

 

[mgarbowski]

The interesting part to me about these graphs is how similar the lines are. We're obviously doing better. You don't need a graph to know that. But we started nearly identical through 8 games, then diverged for 5-6, then went on a somewhat parallel rise.
What we need to hope -- and this seems likely -- is for a second big divergence the rest of the way. The team got worse at the end of last season, earning only 13 points in the last 13 games (including only 4 points in the next 7 games), which was below the season average, which led to lots of discussions of whether the midseason replacements did the team any good. That's an issue we don't really have this year and it seems we're in a nice period of stability.

 

[adam]

View attachment 5333

The interesting part to me about these graphs is how similar the lines are. We're obviously doing better. You don't need a graph to know that. But we started nearly identical through 8 games, then diverged for 5-6, then went on a somewhat parallel rise.
What we need to hope -- and this seems likely -- is for a second big divergence the rest of the way. The team got worse at the end of last season, earning only 13 points in the last 13 games (including only 4 points in the next 7 games), which was below the season average, which led to lots of discussions of whether the midseason replacements did the team any good. That's an issue we don't really have this year and it seems we're in a nice period of stability.

Well we do have to worry about the chemistry of the French Connection. But since they do make a green line in FUT...

 

[SoupInNYC]

This is always a great write-up, but it is particularly enjoyable when the substance of it is so positive!

There are, of course, a ton of permutations and this is MLS, so predicting week to week results is dicey at best plus there is the transfer window, but here's my path to 50 points. I think there is also upside from this and would hope to do better. It feels pretty conservative and supports the point above that we really, really should be making the playoffs.

(1) 7 of 9 from home matches against DC United, Chicago and Columbus - too optimistic given our home record to date? Maybe, but that is a mediocre to bad collection of teams and maybe we've finally figured out this home thing.

(2) 4 of 9 from home matches against Colorado, Dallas and LA - Maybe the three best teams in the league, but not very strong on the road.

(3) 8 from 21 from road matches against New England (2x), Montreal, NJRB, Columbus, Orlando and DC - This feels pretty doable because, while I remain a bit skeptical that we have some unique power on the road, this is a lot of games against a lot of teams that aren't that strong.

(4) 4 from 9 from road matches against Sporting KC, SJ and Houston - I could see us doing worse than this because I still feel our "road warrior" performance is a bit fortunate, but this isn't a great collection of opponents.

After seeing Gotham Gator update his gauntlet to include a tracker, I wanted to revisit this one here to update to what is "needed".

(1) 7 of 9 still - 0 of these games played
(2) 4 of 9 still - 0 of these games played
(3) 2 of 15 - 2 of these games played
(4) 4 of 6 - 1 of these games played

 

[Viewfrom226]

After seeing Gotham Gator update his gauntlet to include a tracker, I wanted to revisit this one here to update to what is "needed".

(1) 7 of 9 still - 0 of these games played
(2) 4 of 9 still - 0 of these games played
(3) 2 of 15 - 2 of these games played
(4) 4 of 6 - 1 of these games played

Thanks! It feels pretty good, no? Even if you still believe that the road form cannot continue, the five games in cat (3) surely should produce more than 2 points!

 

[mgarbowski]

Nice article on MLSoccer about the Blues' road record, adding more info and context to my post on the same topic::
New York City FC continue historic road form in Montreal
Our current 1.90 PPG on the road would be second best in MLS history if we keep it that high. The record is 2.0 by the 2010 LA Galaxy. Four teams share the record of 9 road wins, and three of them won Supporters Shield. Our weak Home record to date very likely keeps us from that achievement even if we do get the 9 road wins. To match LAG's 2.0 PPG, we need 15 points in the last 7 road games, which means either winning 5 or winning 4 with 3 ties and no losses.

 

[Viewfrom226]

Nice article on MLSoccer about the Blues' road record, adding more info and context to my post on the same topic::
New York City FC continue historic road form in Montreal
Our current 1.90 PPG on the road would be second best in MLS history if we keep it that high. The record is 2.0 by the 2010 LA Galaxy. Four teams shre the record of 9 road wins, and three of them won Supporters Shield. Our weka Home record to date very likely keeps us from that achievement even if we do get the 9 road wins. To match LAG's 2.0 PPG, we need 15 points in the last 7 road games, which means either winning 5 or winning 4 with 3 ties and no losses.

I'm legitimately starting to wonder where we can end up. I think we have a very good chance at getting 9 road wins, though I doubt we get to the 2.0 ppg. After this week, it isn't the most impressive collection of road opponents. For this purpose, we have probably been helped by the random allocation of home and away matches against the West.