I'm working on a Grand Unified Theory which I only half-believe but I think it still worth sharing. But I have to finish a couple of work projects first...
2016 Playoff Projections, or Hope By The Numbers
- Thread starter mgarbowski
- Start date
Here's the Grand Unified Theory of Everything, which as noted I'm not sure if I believe it or not. I'm definitely interested in opinions and feedback.
Notes on the above:
1. and 2. These could also be just a small sample and random chance.
3. A small field would probably help all defensive strategies and hurt all offense. This is a game of space. Total Football means making the field bigger when you have the ball and smaller when you don't. But our offense is more focused than ever on waiting for holes to open. There are fewer holes on a small field.
4. Credit to Viewfrom226 for helping me notice this. I also do not know if this ratio is unusual. Most games probably have goal differentials of 0, 1 or 2. But we have none greater than 2 (on the winning side) and the ratio of 1 and 2 seems heavy towards 1 for us. Regardless of whether it is normal or unusual, we have to deal with it.
6. I have to evidence of this but it seems to be a truth universally acknowledged.
7. I also can't say this is true with certainty, but this year we have had:
Toronto - Free kick conceded from wide of the box
Orlando - Saunders tap into Larins face
New England - Free kick deflection off Mendoza
Chicago - free kick conceded - correctly ruled offsides but we were still lucky to get that called
Montreal - Free kick conceded from wide of the box
Vancouver - goal conceded under pressure while playing from the back
Red Bull Wedding - 4 corners and 1 free kick
That's at least one every home game.
- Last year we scored 32 Home goals and 16 Away. This year, with an extra home game so far, we have only 7 Home goals and 11 Away.
- We were shutout only twice all last year at home. This year it has happened 3 times in 7 games.
- Theory: This could be due to our playing from the back on a small field. At home, being patient slow and from the back we are especially susceptible to teams who either park the bus or press. A small field makes both of these easier to play.
- Our 14 wins over 2 seasons include 10 by one goal and 4 by two goals.
- This means we almost never have room for error and are always susceptible to having a win become a tie, or a tie become a loss, based on one fluky, weird, random, goal.
- We are the kings of conceding stupid, fluky goals, and set piece goals.
- This is especially true at home.
- Theory: Small fields favors offense on set pieces and when a team wins the ball on a high press (because it's likely that much closer to goal). Small field hurts offense in regular open play, especially with slow buildup.
- Overall, the small field takes our weaknesses and faults and makes them worse. Combine (1) the fact that we never blow anyone away, and (2) are scoring much less at home this year which amplifies (1), with (3) our tendency to concede on weird plays and set pieces and you have who we are right now.
Notes on the above:
1. and 2. These could also be just a small sample and random chance.
3. A small field would probably help all defensive strategies and hurt all offense. This is a game of space. Total Football means making the field bigger when you have the ball and smaller when you don't. But our offense is more focused than ever on waiting for holes to open. There are fewer holes on a small field.
4. Credit to Viewfrom226 for helping me notice this. I also do not know if this ratio is unusual. Most games probably have goal differentials of 0, 1 or 2. But we have none greater than 2 (on the winning side) and the ratio of 1 and 2 seems heavy towards 1 for us. Regardless of whether it is normal or unusual, we have to deal with it.
6. I have to evidence of this but it seems to be a truth universally acknowledged.
7. I also can't say this is true with certainty, but this year we have had:
Toronto - Free kick conceded from wide of the box
Orlando - Saunders tap into Larins face
New England - Free kick deflection off Mendoza
Chicago - free kick conceded - correctly ruled offsides but we were still lucky to get that called
Montreal - Free kick conceded from wide of the box
Vancouver - goal conceded under pressure while playing from the back
Red Bull Wedding - 4 corners and 1 free kick
That's at least one every home game.
Last season, Matt Besler throw in for a header goal. We just suck balls at defending set pieces.
I don't think we're making the playoffs. I'm ready to say that at the end of May, and even though we're in 4th place.
I hope I'm wrong, but that's my position.
I hope I'm wrong, but that's my position.
I agree. I can't find three teams besides Chicago that will end up worse than us. You'd think Philadelphia but they are the real deal. New England and Columbus are below us now but you have to think they're both going to pick it up. I can't see Toronto slipping out of the playoffs, nor Montreal. Maybe DC? Maybe Orlando?
We need to win at home and we haven't been. Our home games left: RSL, Philadelphia, Red Bulls, Colorado, LA, DC, Dallas, Chicago, Columbus. We need at least 20 points from these.
Currently, bad teams in MLS do reasonably well at home. Since we have been in the league, only 2 teams have a PPG at home below 1.5. Chicago is 1.42 and NYCFC is 1.20. Orlando is 1.54, and everyone else is at 1.6 and above. Playoff teams average 2.0.
You go back a few years and you see some really spectacularly bad Home teams. Toronto and Chivas were just horrible a few years ago. We're not at that level, but right now we are the worst.
You go back a few years and you see some really spectacularly bad Home teams. Toronto and Chivas were just horrible a few years ago. We're not at that level, but right now we are the worst.
In the last 5 seasons of MLS, all 34 game seasons, here is every team that had a .500 road record or better, with their Road Points, Home Points, and Total:
xxxxxxxx Rd Hm Tot
2015 VAN 24 29 53
2014 SKC 25 24 49
2014 SEA 27 37 64
2013 SKC 28 30 58
2012 SKC 29 34 63
2012 SJ 30 36 66
2011 SEA 32 31 63
2011 LAG 26 41 67
Here's what I notice. Outside of 2014 SKC, they are all really good teams. Vancouver 2015 has the next lowest point total and it finished 3rd in Supporters shield. Also, 6 of the 8 had more Home points than Road Points. For teams like us, just looking to sneak into the playoffs somehow, 2014 SKC is our only model. How we get there:
Here is our current Home/Road Point Record Split:
Home 1-2-5 8pts
Away 3-2-1 10 Pts
Here's how we get to 2014 SKC totals:
Home: 5-3-1 16 Pts 1.78 PPG
Road 4-4-3 15 Pts 1.36 PPG
You can adjust that. Maybe you want to flip a win between Home and Away. If you think it will take fewer than 49 points to make the playoffs, play with it that way. I think there's a decent chance that 46 or 47 makes it, but there's still not a lot of wiggle room.
If NYCFC doesn't manage to keep up it's Away success, and slips to just one game under on the Road for the year, here is what we are looking at as an example to get to 49 Points.
Home 6-2-1 19 Pts 2.11 PPG
Away 3-5-3 12 Pts 1.091 PPG
Current PPG Standings:
So the current Playoff line looks like 42 points. If that sounds encouraging remember that it started May at 38.
Ignoring Home/Away here is what we need to do to reach assorted point totals:
xxxxxxxx Rd Hm Tot
2015 VAN 24 29 53
2014 SKC 25 24 49
2014 SEA 27 37 64
2013 SKC 28 30 58
2012 SKC 29 34 63
2012 SJ 30 36 66
2011 SEA 32 31 63
2011 LAG 26 41 67
Here's what I notice. Outside of 2014 SKC, they are all really good teams. Vancouver 2015 has the next lowest point total and it finished 3rd in Supporters shield. Also, 6 of the 8 had more Home points than Road Points. For teams like us, just looking to sneak into the playoffs somehow, 2014 SKC is our only model. How we get there:
Here is our current Home/Road Point Record Split:
Home 1-2-5 8pts
Away 3-2-1 10 Pts
Here's how we get to 2014 SKC totals:
Home: 5-3-1 16 Pts 1.78 PPG
Road 4-4-3 15 Pts 1.36 PPG
You can adjust that. Maybe you want to flip a win between Home and Away. If you think it will take fewer than 49 points to make the playoffs, play with it that way. I think there's a decent chance that 46 or 47 makes it, but there's still not a lot of wiggle room.
If NYCFC doesn't manage to keep up it's Away success, and slips to just one game under on the Road for the year, here is what we are looking at as an example to get to 49 Points.
Home 6-2-1 19 Pts 2.11 PPG
Away 3-5-3 12 Pts 1.091 PPG
Current PPG Standings:
So the current Playoff line looks like 42 points. If that sounds encouraging remember that it started May at 38.
Ignoring Home/Away here is what we need to do to reach assorted point totals:
ESPN put a stat up during the game yesterday that said no team with less than 6 home wins in a season has ever made the playoffs. We have 1 win in 8 at home so far
Let's make history! We seem to be good at doing that! So far it has been in a bad way but we make history none the less!
Last year the team was 17th in league-wide Supporters Shield race, but 19th in points won at home.
This year NYCFC is currently 10th for Supporters Shield, and 20th in home record (TOR has fewer points but better PPG and many fewer GP).
This year NYCFC is currently 10th for Supporters Shield, and 20th in home record (TOR has fewer points but better PPG and many fewer GP).
It's worth mentioning that we're first in Away record also, so 10th overall is right where we should be I suppose.
Right. And the correct assumption is that our home record will improve over the season while our away record comes down to earth, leaving us about where we are now overall - i.e. in the playoffs. It is a real reach to assume our away record gets worse and home stays the same.
There are no immutable laws of science that say a team has to have a certain home record or away record to make the playoffs. There are trends that have occurred in the past, but that's correlation not causality. Its weak to try to predict our fall based on a home record.
There are plenty of other reasons we might miss playoffs, but (and not targeting mgarbowski, I've seen many other predicting this) but reaching some magic home record should not be one of them.
There are plenty of other reasons we might miss playoffs, but (and not targeting mgarbowski, I've seen many other predicting this) but reaching some magic home record should not be one of them.
The problem is the Mean for Home Games for playoff teams is double our current rate. To do that over the remaining games means winning 6 of the remaining 9 home games. I find assuming that to be tough, so we're left guessing how close we get to that and how far the Away record drops. Here is a scenario that works:
Home 5-3-1 16 Pts 1.78 PPG
Away 4-4-3 15 Pts 1.36 PPG
Those 31 points added to the current 18 gets us to 49. One way or another you have to figure we need to win at least 9 of our remaining 20 games. It could require more.
Think of that as 4.5 wins per 10 games.
Last year we earned 2.94 Wins per 10 games.
Last year in the last 21 games after the winless streak we managed 4.28 Wins per 10 games.
This year we are sitting at 2.86 Wins per 10 games.
Basically to hit the playoffs we likely need to do a bit better than we did after the winless streak last year. It's not out of reach. But it's also something we've never done. Both are true, which is one of the reasons it's hard to make predictions, especially about the future.
You're completely right. I'm not blind to this. But when making predictions in sports you can go 2 ways:
1. You can make wild predictions of unusual events that almost never come true, and then when 1 in 1,000 does, beat your chest and say "I predicted that."
2. Or you can look for trends and predict that things will go the way they almost always go. You'll never predict the rare Black Swan event, but you'lee get a lot more right and be more useful.
Leicester was properly at 5,000 to 1 last year, and came through. But if you predict that the team with the longest odds in the top league of English football won't be champions for 100 years, you'd be right 99 times. You could even be right 999 times in 1,000. I'll take that.
If you want to predict NYCFC will set new standards for H/A records and make the playoffs because of what you've seen in our play, that's great and please explain why.
If you want to predict NYCFC will set new standards for H/A records and make the playoffs because you're an optimist, good for you.
If you simply want to assume current trends continue, that's dubious.
I'm still going to predict based on the odds, while hoping for the best.
I still think this is overblown.
Imagine you have two weighted coins, one which comes up heads 2/3rds of the time (home), the other 1/3rd (away).
Now, after tossing each coin 7 times say you've gotten heads 7 times, as expected, but oddly enough you've gotten 5 successful tosses from the "away" coin and only two from the "home" coin.
Should you be concerned about coming up with enough heads after 20 more tosses? After all, no one has ever done it before! Of course not. All that matters is whether your total results put you where you ought to be.
Now, of course the difference here in the real world is that we don't know the "true" probability of winning at home vs away. Sticking with the metaphor, we might start to wonder if our coins aren't weighted the way we initially assumed.
I think this neatly encapsulates the question at hand.
But we have more data than watching a coin. We saw the games! Are the results we've seen more consistent with a team that has been fortunate away from home and unfortunate at home, or a team that truly performs better away than at home?
And to answer that we have to fall back on the eye test. I still say that based on our play, I'm not too concerned.
We have dominated in 4 of our home games (Vancouver, Orlando, Montreal, and Orlando). We were dominated in Portland. I expect the results to normalize over he rest of the year. I'm far more concerned by the increasing strength of our opponents and put failure to win against mediocre opponents regardless of location than the home/away record.
Imagine you have two weighted coins, one which comes up heads 2/3rds of the time (home), the other 1/3rd (away).
Now, after tossing each coin 7 times say you've gotten heads 7 times, as expected, but oddly enough you've gotten 5 successful tosses from the "away" coin and only two from the "home" coin.
Should you be concerned about coming up with enough heads after 20 more tosses? After all, no one has ever done it before! Of course not. All that matters is whether your total results put you where you ought to be.
Now, of course the difference here in the real world is that we don't know the "true" probability of winning at home vs away. Sticking with the metaphor, we might start to wonder if our coins aren't weighted the way we initially assumed.
I think this neatly encapsulates the question at hand.
But we have more data than watching a coin. We saw the games! Are the results we've seen more consistent with a team that has been fortunate away from home and unfortunate at home, or a team that truly performs better away than at home?
And to answer that we have to fall back on the eye test. I still say that based on our play, I'm not too concerned.
We have dominated in 4 of our home games (Vancouver, Orlando, Montreal, and Orlando). We were dominated in Portland. I expect the results to normalize over he rest of the year. I'm far more concerned by the increasing strength of our opponents and put failure to win against mediocre opponents regardless of location than the home/away record.
Two things: Leicester should never have been 5000 to 1 not because they were underrated but because no underdog should ever have odds that low.
I am not arguing about whether we should predict based on odds. The question is which odds to use. I maintain that total points now compared to expected points necessary is more relevant than the exact number of home and away games won. As it happens we're off the pace even so - but by a small amount, not by a large one. I don't think that NYCFC averaging 2 points per home game and 1 point per away game for the rest of the season and making he playoffs would qualify as a black swan event.
Similar threads
- Poll
