Points from remaining game

[spyinthesky]

A mate of mine shared some information on guessing point predictions based on the following detail which estimates % chance of getting a win from individual games :~
FIXTURE 
                         Total Pts.    Chances of Winning.    Est Pts.                                  Total Pts        Chances of Winning.         Est Pts     
Man City (H)                                                                                   Burnley (A)              3                            40%                                    1.2.
Spurs       (A)
Man Utd   (A)                                                                                  Brentford (A)            3                            50%                                     1.5
Chelsea   (H)
Liverpool (H)                                                                                    Everton   (H)     
                             15.                      10%.                                 1.5.            Norwich  (H)
A Villa       (A)                                                                                    Watford   (H) 
Leeds       (A)                                                                                    C Palace  (H)
Arsenal     (H)                                                                                    Newcastle (H)
Brighton   (A)                                                                                                                     15                           60%                                   9
Leicester  (A)
                            15.                      30%                                  4.5

Games                               17
Points Available                51
Expected Points               17.7            (Average 1.04 per game)
Current Points                  24              (Average 1.14 per game)
Estimated Final Points      41/42

On this basis we should be safe for another year!!                                                                                                      

[Saint Garrett]

Who decides the % likelihood of us winning?

[CB Fry]

Recent results have made me much more confident - we need about 12 points from 17 games, so even a horrendous drop in form to Norwich/Watford levels should see us stay up. By the time we play Leeds in April I'd hope we've got most of what we need.

The final tally feels like it is going to be pretty low this year. Newcastle staying up with 33 points or something.

[Toussaint]

8 minutes ago, Saint Garrett said:

Who decides the % likelihood of us winning?

TWAR

[Charlie Wayman]

Seems like an exercise in futility and despair. Crystal balls are a myth.

[SaintBobby]

The odd thing about this "methodology" is that the "chance of winning" and "estimated points" don't correlate.

My understanding is that, in the Premier League, you get one point for drawing a match. The calculation on this thread assumes no draws (or no points for drawing).

If you have a 50% chance of winning a given game, your estimated points return is higher than 1.5. You presumably have some sort of chance (25%?) of drawing it, which needs to be added. (in this case, it would mean your est, points is 1.75).

If you have a 10% chance of winning a match, you also presumable have some chance of drawing it. (15%?). In this case your est. points goes from 0.3 to 0.45.

This pretty glaring flaw explains why the original poster only expects us to get 1.04 points per game - he has ruled out Saints getting any further points at all from draws.

 

[Cartman]

12 hours ago, spyinthesky said:

On this basis we should be safe for another year!!                                                                                                      

Jumping to conclusions there. It's not like we have almost double the amount of points as the team in 18th and that there are 4 teams scoring so few points that they will be lucky to reach 30.

[Cartman]

10 hours ago, Charlie Wayman said:

Seems like an exercise in futility and despair. Crystal balls are a meth.

That's how I read it at first.

[CB Fry]

5 hours ago, SaintBobby said:

The odd thing about this "methodology" is that the "chance of winning" and "estimated points" don't correlate.

My understanding is that, in the Premier League, you get one point for drawing a match. The calculation on this thread assumes no draws (or no points for drawing).

If you have a 50% chance of winning a given game, your estimated points return is higher than 1.5. You presumably have some sort of chance (25%?) of drawing it, which needs to be added. (in this case, it would mean your est, points is 1.75).

If you have a 10% chance of winning a match, you also presumable have some chance of drawing it. (15%?). In this case your est. points goes from 0.3 to 0.45.

This pretty glaring flaw explains why the original poster only expects us to get 1.04 points per game - he has ruled out Saints getting any further points at all from draws.

 

But it doesn't do that. For a start the estimated final points tally around 17 points which is an impossibility to achieve with wins only and no draws.

Secondly it clearly clusters games by difficulty, and assumes in some cases we will get a point or two from some clusters of games. Which can only be achieved by drawing them. For example: one and a half points from Man C/Man U/Chelsea/Spurs/Liverpool suggests we, at best, will draw two, therefore winning none. Which feels fair enough.

So it's not really thinking about wins and draws per se, its just doing the old SaintsWeb favourite of "I reckon we'll get 5 points from the next six games" routine and extrapolating it across half a season.

It doesn't actually say we are going to win or draw any specific match.

Edited 21 January, 2022 by CB Fry

[SaintBobby]

On 21/01/2022 at 06:51, CB Fry said:

But it doesn't do that. For a start the estimated final points tally around 17 points which is an impossibility to achieve with wins only and no draws.

Secondly it clearly clusters games by difficulty, and assumes in some cases we will get a point or two from some clusters of games. Which can only be achieved by drawing them. For example: one and a half points from Man C/Man U/Chelsea/Spurs/Liverpool suggests we, at best, will draw two, therefore winning none. Which feels fair enough.

So it's not really thinking about wins and draws per se, its just doing the old SaintsWeb favourite of "I reckon we'll get 5 points from the next six games" routine and extrapolating it across half a season.

It doesn't actually say we are going to win or draw any specific match.

Well, from what I can see, the expected points are based on multiplying "chance of wining" by 3 points per match. 

So, he reckons we have a 10% chance of winning in those 5 most dfficult matches and so multiples the 15 points available by 0.1 and comes up with 1.5. Presumably giving a zero chance of drawing any of them, which is weird.

You're right that it's impossible to get 17 points just from wins, of course - but that doesn't really matter mathematically. Although from each match you can only collect 3, 1 or zero points, you could put the expected value of a match at something weird like 1.856 points. You're not actually going to get 1.856 points, it just indicates where you are on the spectrum from 0-3. In that scenario, a draw is a bit of a disapppointment, a win pretty good and a defeat a serious disapppointment. Against City yesterday, if we were reckoned to be 10% to win and 20% to draw, that comes out as an expected 0.5 points - so a draw would have exceeded expectations.

What the OP's model does is to only consider the points you can anticipate geting from winning. It's fair to say he expects us to get c17 points from winning games (although, obviously the actual number would be to get 15 or 18 or whatever) but, weirdly, no points from draws.

Most betting/predicting models at the start of the season will spit something out like "Saints expected points total is 42.76". They don't actually think we will get that weird 0.76 - it's just a number that gives you the predicted best chance of getting as close to the final tally as you can (even though it's certain you'll be wrong by at least 0.24 points).

 

[tajjuk]

Personally I think we are getting better and more solid, barring another horrible injury period (which we have more depth to cope with) and reckon we'll be nearer 50 points or so

[CB Fry]

1 hour ago, SaintBobby said:

Well, from what I can see, the expected points are based on multiplying "chance of wining" by 3 points per match. 

So, he reckons we have a 10% chance of winning in those 5 most dfficult matches and so multiples the 15 points available by 0.1 and comes up with 1.5. Presumably giving a zero chance of drawing any of them, which is weird.

You're right that it's impossible to get 17 points just from wins, of course - but that doesn't really matter mathematically. Although from each match you can only collect 3, 1 or zero points, you could put the expected value of a match at something weird like 1.856 points. You're not actually going to get 1.856 points, it just indicates where you are on the spectrum from 0-3. In that scenario, a draw is a bit of a disapppointment, a win pretty good and a defeat a serious disapppointment. Against City yesterday, if we were reckoned to be 10% to win and 20% to draw, that comes out as an expected 0.5 points - so a draw would have exceeded expectations.

What the OP's model does is to only consider the points you can anticipate geting from winning. It's fair to say he expects us to get c17 points from winning games (although, obviously the actual number would be to get 15 or 18 or whatever) but, weirdly, no points from draws.

Most betting/predicting models at the start of the season will spit something out like "Saints expected points total is 42.76". They don't actually think we will get that weird 0.76 - it's just a number that gives you the predicted best chance of getting as close to the final tally as you can (even though it's certain you'll be wrong by at least 0.24 points).

 

Sorry this makes no sense to me. If a predictor looks at say 5 tough games and comes back with a forecast of 1.5 points for those games then that means in reality that the forecaster thinks we are likely to draw one or maybe two of those five games. You don't need a separate and additional "likelihood of a draw" calculator on top. Just utterly pointless when the work is already done: five tough games, 1.5 points: we probably won't win any but we might draw one or two. Done.

Pretty simple and obvious to me. I genuinely don't understand what you don't understand. 

Edited 23 January, 2022 by CB Fry

[SNSUN]

By my guesstimation we should get somewhere between 20 and 25 points for the rest of the season. We’re a bit Jekyll and Hyde but I can’t see a collapse like last season. Top half? Doubtful, but around 12th/13th again I reckon. Then we’ll see if the new owners can or can’t invest a bit in the summer.

 

Someone will slip towards relegation that isn’t there currently, possibly Brentford or even Everton, but ultimately I still think it’ll be Burnley, Norwich and Watford for the drop. Hard to see past them.

[SaintBobby]

13 minutes ago, CB Fry said:

Sorry this makes no sense to me. If a predictor looks at say 5 tough games and comes back with a forecast of 1.5 points for those games then that means in reality that the forecaster thinks we are likely to draw one or maybe two of those five games. You don't need a separate and additional "likelihood of a draw" calculator on top. Just utterly pointless when the work is already done: five tough games, 1.5 points: we probably won't win any but we might draw one or two. Done.

Pretty simple and obvious to me. I genuinely don't understand what you don't understand. 

If the calculation is that we have a 50% of winning a given match (say, Brentford away), then the expected points from that game is higher than 1.5. You have a 50% chance of 3 points (=1.5) and, on top of that, some chance of 1 point (maybe 25%, so +0.25? for a total of 1.75)

If we have a 10% chance of winning each of those 5 listed tough games, you would expect to get more than 1.5 points from them in total (ignoring, of course, that we have already take a point from one of those matches yesterday).

I don't know if the "chances of winning" precentages that he lists are right. But if they are right, we should expect to get more than the 17.7 points he calculates from those matches (again, ignoring that we've already taken 1 from Man City). 

[CB Fry]

5 minutes ago, SaintBobby said:

If the calculation is that we have a 50% of winning a given match (say, Brentford away), then the expected points from that game is higher than 1.5. You have a 50% chance of 3 points (=1.5) and, on top of that, some chance of 1 point (maybe 25%, so +0.25? for a total of 1.75)

If we have a 10% chance of winning each of those 5 listed tough games, you would expect to get more than 1.5 points from them in total (ignoring, of course, that we have already take a point from one of those matches yesterday).

I don't know if the "chances of winning" precentages that he lists are right. But if they are right, we should expect to get more than the 17.7 points he calculates from those matches (again, ignoring that we've already taken 1 from Man City). 

I think you are just massively massively over thinking it.

5 difficult games, 15 points up for grabs, but predict to get maybe one or two points - ie two draws at best. I don't see what you doing Pythagoras routine is adding to that prediction - it's a perfectly sound guess - 15 points but we'll likely get 10% of those points. Makes perfect sense.

For the other end of the scale - the "easier games" - again 15 points up for grabs, but it make an assumption we'll get 9 points from them. 3 wins. Maybe two wins, three draws. That ballpark, 60% of the points available. Again, makes perfect sense. 

So far this season (including yesterday now) we've got 38% of the available points, this predictor (created before yesterday) assumes we'd get 35% of the remaining points. 

So if anything it is on the pessimistic side, but it's a perfectly sound set of predictions, and even with this pessimistic prediction it bottoms out that we are going to stay up. 

All makes perfect sense. 

[SaintBobby]

24 minutes ago, CB Fry said:

I think you are just massively massively over thinking it.

5 difficult games, 15 points up for grabs, but predict to get maybe one or two points - ie two draws at best. I don't see what you doing Pythagoras routine is adding to that prediction - it's a perfectly sound guess - 15 points but we'll likely get 10% of those points. Makes perfect sense.

For the other end of the scale - the "easier games" - again 15 points up for grabs, but it make an assumption we'll get 9 points from them. 3 wins. Maybe two wins, three draws. That ballpark, 60% of the points available. Again, makes perfect sense. 

So far this season (including yesterday now) we've got 38% of the available points, this predictor (created before yesterday) assumes we'd get 35% of the remaining points. 

So if anything it is on the pessimistic side, but it's a perfectly sound set of predictions, and even with this pessimistic prediction it bottoms out that we are going to stay up. 

All makes perfect sense. 

Well, I don't think there's much to think. If the "chance of winning" (which is what it says) is as he says, the likely points haul is higher. You might think his "points prediction" is right...or his "chances of winning" predictions for the matches.  But one thing is for sure, they can't both be right.

Overall, you'd probably also expect us to take a higher % of points from the remaining games than the previous ones, FWIW. (given we've played City, West Ham and Wolves twice and these would probably be lower than average on predicted points haul).

If the OP had said "we'll probably get 18 or so points from the remaining matches as a hunch", well he's probably not a million miles off. But he's actually put forward a mathematical method of getting there which doesn't add up.

[CB Fry]

3 minutes ago, SaintBobby said:

Well, I don't think there's much to think. If the "chance of winning" (which is what it says) is as he says, the likely points haul is higher. You might think his "points prediction" is right...or his "chances of winning" predictions for the matches.  But one thing is for sure, they can't both be right.

Overall, you'd probably also expect us to take a higher % of points from the remaining games than the previous ones, FWIW. (given we've played City, West Ham and Wolves twice and these would probably be lower than average on predicted points haul).

If the OP had said "we'll probably get 18 or so points from the remaining matches as a hunch", well he's probably not a million miles off. But he's actually put forward a mathematical method of getting there which doesn't add up.

Like I say, you're massively over thinking it.

[SaintBobby]

Well, I’m not quite sure how pointing out how his “chance of winning” doesn’t lead to the predicted points outcome he suggests is “massively over-thinking it”.

It’s hardly quantum physics is it?…it’s not even up the rigours of GCSE maths….

[Turkish]

We just need to get to 40 points as soon as possible or we are definitely going down 

[Charlie Wayman]

We need 13 points from sixteen games to be safe. How difficult is that?

Let's focus on how to improve our squad in the summer than worry about what's left of this season

[CB Fry]

50 minutes ago, SaintBobby said:

Well, I’m not quite sure how pointing out how his “chance of winning” doesn’t lead to the predicted points outcome he suggests is “massively over-thinking it”.

It’s hardly quantum physics is it?…it’s not even up the rigours of GCSE maths….

Its not me that looks at a prediction of one and a half points from 5 tough games and interprets it as "this person doesn't know how many points you get from a draw".

Really not that complicated.

[woodsaint1]

20 minutes ago, Charlie Wayman said:

We need 13 points from sixteen games to be safe. How difficult is that?

Let's focus on how to improve our squad in the summer than worry about what's left of this season

We dont even need 13 points, no chance one of those bottom 3/4 teams are reaching 38 points (they would likely need 39 with their GD). 3 more wins and we're safe. We'll have that by early/mid March

[Turkish]

These threads are always fascinating 

[Sarisbury Saint]

 😴 

[Saint_clark]

I looked through the fixtures at work today and I had us finishing on 45 points. Felt like I was being conservative as well. 

[Convict Colony]

On 20/01/2022 at 16:27, spyinthesky said:

A mate of mine shared some information on guessing point predictions based on the following detail which estimates % chance of getting a win from individual games :~
FIXTURE 
                         Total Pts.    Chances of Winning.    Est Pts.                                  Total Pts        Chances of Winning.         Est Pts     
Man City (H)                                                                                   Burnley (A)              3                            40%                                    1.2.
Spurs       (A)
Man Utd   (A)                                                                                  Brentford (A)            3                            50%                                     1.5
Chelsea   (H)
Liverpool (H)                                                                                    Everton   (H)     
                             15.                      10%.                                 1.5.            Norwich  (H)
A Villa       (A)                                                                                    Watford   (H) 
Leeds       (A)                                                                                    C Palace  (H)
Arsenal     (H)                                                                                    Newcastle (H)
Brighton   (A)                                                                                                                     15                           60%                                   9
Leicester  (A)
                            15.                      30%                                  4.5

Games                               17
Points Available                51
Expected Points               17.7            (Average 1.04 per game)
Current Points                  24              (Average 1.14 per game)
Estimated Final Points      41/42

On this basis we should be safe for another year!!                                                                                                      

Come on bro if you learnt anything from Ras's Ted talk it's goal difference and also shot for v against that help with predicting results/finish.

Bookies know best.

[SaintBobby]

1 hour ago, CB Fry said:

Its not me that looks at a prediction of one and a half points from 5 tough games and interprets it as "this person doesn't know how many points you get from a draw".

Really not that complicated.

It’s pretty simple isn’t it? You have a chance to win games and a chance to draw too. You’re massively over-thinking this.

[CB Fry]

48 minutes ago, SaintBobby said:

It’s pretty simple isn’t it? You have a chance to win games and a chance to draw too. You’re massively over-thinking this.

Well done on getting there in the end. I'm proud of you.

[SaintBobby]

23 minutes ago, CB Fry said:

Well done on getting there in the end. I'm proud of you.

So…we will likely get more than 17.7 points. Because we will win some games and draw some others.

Glad it’s falling into place for you. You were massively over-thinking it, but bit by painful bit you got there in the end.

Well done you 👍👍👍

[CB Fry]

7 hours ago, SaintBobby said:

So…we will likely get more than 17.7 points. Because we will win some games and draw some others.

Glad it’s falling into place for you. You were massively over-thinking it, but bit by painful bit you got there in the end.

Well done you 👍👍👍

All there on the thread. Some of us understood it at first glance.

Others, let's call them you, started blathering on about how someone predicting one or two points from five games doesn't, er, know how many points you get for a draw. Or, when the phrase "available points" is used, it already includes points available for a draw.

Bless x x

Edited 24 January, 2022 by CB Fry

[SaintBobby]

2 hours ago, CB Fry said:

All there on the thread. Some of us understood it at first glance.

Others, let's call them you, started blathering on about how someone predicting one or two points from five games doesn't, er, know how many points you get for a draw. Or, when the phrase "available points" is used, it already includes points available for a draw.

Bless x x

Bless. But it’s impossible to “understand” at “first glance” and, sadly, you now seem to have failed to understand it again. So sad to witness this regression.

I’ll amend the original post, maybe that way you’ll find it easier to grasp. It’s really not as difficult as you’re making it out to be.

[SaintBobby]

On 20/01/2022 at 15:27, spyinthesky said:

A mate of mine shared some information on guessing point predictions based on the following detail which estimates % chance of getting a win from individual games :~
FIXTURE 
                         Total Pts.    Chances of Winning.    Est Pts.                                  Total Pts        Chances of Winning.         Est Pts     
Man City (H)                                                                                   Burnley (A)              3                            40%                                    1.2.
Spurs       (A)
Man Utd   (A)                                                                                  Brentford (A)            3                            50%                                     1.5
Chelsea   (H)
Liverpool (H)                                                                                    Everton   (H)     
                             15.                      10%.                                 1.5.            Norwich  (H)
A Villa       (A)                                                                                    Watford   (H) 
Leeds       (A)                                                                                    C Palace  (H)
Arsenal     (H)                                                                                    Newcastle (H)
Brighton   (A)                                                                                                                     15                           60%                                   9
Leicester  (A)
                            15.                      30%                                  4.5

Games                               17
Points Available                51
Expected Points FROM WINS              17.7            (Average 1.04 per game)
Current Points                  24              (Average 1.14 per game)

THE SMARTER AMONGST YOU MIGHT HAVE WORKED OUT THAT YOU CAN'T ACTUALLY ACCRUE 17.7 POINTS FROM WINNING GAMES - WINS LEAD TO 3 POINTS AND SO THE FINAL POINTS TOTAL FROM WINNING GAMES WILL ACTUALLY BE A MULTIPLE OF THREE. HOWEVER, APPLYING THE "CHANCE OF WINNING" PERCENTAGE TO ALL THE MATCHES ABOVE AVERAGES OUT AT 17.7 - MEANING THE BEST GUESS IS THAT WE WILL PROBABLY WIN 5 OR 6 OF OUR REMAINING GAMES

ALSO, FOR THOSE OF YOU WHO HAVE FOLLOWED SAINTS THIS SEASON, YOU'LL NOTICE THAT MATCHES CAN ACTUALLY END IN DRAWS. IN FACT, A SURPRISINGLY HIGH NUMBER OF OUR LEAGUE GAMES THIS SEASON HAVE DONE SO (NOW TEN TO BE EXACT, FOR THOSE COUNTING). YOU GET 1 POINT FOR EACH DRAW. WE WILL PROBABLY DRAW SOME OF THE MATCHES LISTED ABOVE - ALTHOUGH IT'S UNLIKELY TO BE QUITE AS HIGH AS IT HAS BEEN TO DATE.

LET'S SAY, IN ORDER NOT TO "MASSIVELY OVERTHINK" THINGS THAT THERE'S A 25% CHANCE OF EACH GAME ENDING IN A DRAW. THAT MEANS WE CAN ADD 0.25 POINTS FOR EACH OF THE REMAINING 17 GAMES.

EXPECTED POINTS FROM DRAWS 4.25 (AVERAGE OF 0.25 PER GAME)

EXPECTED POINTS FROM BOTH WINS AND DRAWS 21.95 (1.29 POINTS PER GAME)

Estimated Final Points      45.95 (YOU CAN ROUND THIS UP TO 46 OR MAYBE EVEN CLOSE TO 47 GIVEN WE HAVE NOW SECURED A POINT AGAINST MAN CITY, WHERE WE WERE PROBABLY ONLY EXPECTING ABOUT "HALF A POINT")

On this basis we should be safe for another year!!                                                                                                    

Hope this helps the mathematically hard of thinking. 

Edited 24 January, 2022 by SaintBobby

[CB Fry]

8 minutes ago, SaintBobby said:

Hope this helps the mathematically hard of thinking. 

At least you're taking it well.

[SaintBobby]

7 minutes ago, CB Fry said:

At least you're taking it well.

Always happy to help those who struggle with multiplication and addition. Doesn’t come easy to some people, so it’s usually best to talk them through step by step.

[CB Fry]

1 hour ago, SaintBobby said:

Always happy to help those who struggle with multiplication and addition. Doesn’t come easy to some people, so it’s usually best to talk them through step by step.

Needs more bold font and CAPITALS and underlines sweetheart.

[SaintBobby]

6 minutes ago, CB Fry said:

Needs more bold font and CAPITALS and underlines sweetheart.

Nah. Not really, hun. Hope it’s helped you work through the basic maths though, sweetie.

[CB Fry]

3 minutes ago, SaintBobby said:

Nah. Not really, hun. Hope it’s helped you work through the basic maths though, sweetie.

I don't need to read whatever it is you have rage-typed, as I understood the opening post of the thread, at first glance, right from the off. You've admitted that you didn't (apparently it's "impossible") but that's your issue not mine.

[SaintBobby]

29 minutes ago, CB Fry said:

I don't need to read whatever it is you have rage-typed, as I understood the opening post of the thread, at first glance, right from the off. You've admitted that you didn't (apparently it's "impossible") but that's your issue not mine.

No, you didn’t really understand it. Certainly, you didn’t grasp the big flaw in it. You couldn’t understand that we will accrue points both from wins and also from draws, but that it only showed points likely to come from winning games, so the working method was clearly wrong (or perhaps you just didn’t read it properly!)

Apparently, noting that we might pick up some points from draws between now and the end of the season was “massively over thinking” things. 😂😂😂

I’m pleased to have had the opportunity to put you right, though. All part of my commitment to care in the community.

[Noodles34]

17 hours ago, Charlie Wayman said:

We need 13 points from sixteen games to be safe. How difficult is that?

Let's focus on how to improve our squad in the summer than worry about what's left of this season

Have a look at our last 16 games last season

 

[CB Fry]

3 hours ago, SaintBobby said:

No, you didn’t really understand it. Certainly, you didn’t grasp the big flaw in it. You couldn’t understand that we will accrue points both from wins and also from draws, but that it only showed points likely to come from winning games, so the working method was clearly wrong (or perhaps you just didn’t read it properly!)

Apparently, noting that we might pick up some points from draws between now and the end of the season was “massively over thinking” things. 😂😂😂

I’m pleased to have had the opportunity to put you right, though. All part of my commitment to care in the community.

You can't get 1.5 points from winning a game. There's a five game cluster with 1.5 points attached to it, which clearly shows a prediction that we will draw one or two of those games and win none. 

You basically become massively hung up on the phrase "chance of winning". All it actually is a proxy to get to an assumption that we will pick up a % of points from groups of games - 10% from hard games, 60% from easier games - so 1.5 points here, 9 points there. We won't win all games, we won't lose all the ones we won't win, we will draw some. Hence 1.5 points in some cases, hence 17 points as a working assumption which is impossible to achieve by just wins.

Over thinking it, massively, and then writing tantrums in bold UNDERLINED IN CAPITALS....absolutely laughable

I got it in post one, in seconds. You are still raging and raging and raging.

You continue to embarrass yourself. Personally I am looking forward to the mods splitting this off into its own thread. 

 

[SaintBobby]

4 minutes ago, CB Fry said:

You can't get 1.5 points from winning a game. There's a five game cluster with 1.5 points attached to it, which clearly shows a prediction that we will draw one or two of those games and win none. 

You basically become massively hung up on the phrase "chance of winning". All it actually is a proxy to get to an assumption that we will pick up a % of points from groups of games - 10% from hard games, 60% from easier games - so 1.5 points here, 9 points there. We won't win all games, we won't lose all the ones we won't win, we will draw some. Hence 1.5 points in some cases, hence 17 points as a working assumption which is impossible to achieve by just wins.

Over thinking it, massively, and then writing tantrums in bold UNDERLINED IN CAPITALS....absolutely laughable

I got it in post one, in seconds. You are still raging and raging and raging.

You continue to embarrass yourself. Personally I am looking forward to the mods splitting this off into its own thread. 

 

😂😂😂😂 omfg you still don’t get it. It’s hilarious.

Predicting an expected return of 1.5 points from 5 particular games doesn’t mean 1 or 2 draws. The OP says it - in black and white himself himself - it’s 5 long shots at a win. You’ll probably win 0 or 1 of them. 1.5 is an average guess of how many points wins will deliver you from those games. The clue is kind of there…when he writes the words “chances of winning”.

You seem to be “massively over- thinking it”. To most people the phrase “chances of winning” is pretty straightforward.

It means the chance you have of winning that game. So, for example, a “chance of winning” of 50% means you have a 50-50 chance of winning that game. Your anticipated points return from actually winning that game is 1.5. It’s not 1.5 because you think you will draw that game one and a half times.

To you, for whatever reason, the phrase “chances of winning” is apparently defined as meaning “chances of drawing”.

Bizarre. Kind of cute and rather endearing. But truly bizarre. Bless. 😂😂😂

[CB Fry]

23 minutes ago, SaintBobby said:

😂😂😂😂 omfg you still don’t get it. It’s hilarious.

Predicting an expected return of 1.5 points from 5 particular games doesn’t mean 1 or 2 draws. The OP says it - in black and white himself himself - it’s 5 long shots at a win. You’ll probably win 0 or 1 of them. 1.5 is an average guess of how many points wins will deliver you from those games. The clue is kind of there…when he writes the words “chances of winning”.

You seem to be “massively over- thinking it”. To most people the phrase “chances of winning” is pretty straightforward.

It means the chance you have of winning that game. So, for example, a “chance of winning” of 50% means you have a 50-50 chance of winning that game. Your anticipated points return from actually winning that game is 1.5. It’s not 1.5 because you think you will draw that game one and a half times.

To you, for whatever reason, the phrase “chances of winning” is apparently defined as meaning “chances of drawing”.

Bizarre. Kind of cute and rather endearing. But truly bizarre. Bless. 😂😂😂

"Chance of drawing" isn't a thing you absolute f**king plank.

15 available points from five tough games, a prediction we'll get 10% of those points, so a point and a half therefore a draw at worst, two draws at best.

You don't need a "chance of drawing" calculator on top. 

Points gained from draws are just intrinsic in the model. It's there.

Absolutely laughable that you are rage typing in bold and underlined about something so basic.

You're also making things up from the opening post now: the phrase "five long shots at a win" is not there.

Honestly give it up. Embarrassing.

 

 

[OttawaSaint]

Is this how maths geeks have sex?

*David Attenborough voice*

The math geeks will now engage in a thrillingly intricate performance of football probability calculations to see who goes on top...

[Charlie Wayman]

3 hours ago, Noodles34 said:

Have a look at our last 16 games last season

 

We have a different team this season

[SaintBobby]

8 minutes ago, CB Fry said:

"Chance of drawing" isn't a thing you absolute f**king plank.

15 available points from five tough games, a prediction we'll get 10% of those points, so a point and a half therefore a draw at worst, two draws at best.

You don't need a "chance of drawing" calculator on top. 

Points gained from draws are just intrinsic in the model. It's there.

Absolutely laughable that you are rage typing in bold and underlined about something so basic.

You're also making things up from the opening post now: the phrase "five long shots at a win" is not there.

Honestly give it up. Embarrassing.

 

 

Can you read? He isn’t predicting we will take 10% of the points from those tough games. He is predicting we have a 10% “chance of winning” each of them.

If you have a really tough run of ten games, where you have a 10% “chance of winning” in each case, how many points would you guess we’d take from those games?

(A) 3 points?

Or (B) More than 3 points?

(I’ll give you a clue. It’s B).

So, and this requires some complex division now, if you have a 10% “chance of winning” in a run of just five (rather than ten) tough games what’s the expected points value of those games (to as many decimal places as you like as it’s an expected average value)?

Is it:

(A) 1.5 points?

or (B) more than 1.5 points?

(I’ll give you a clue - it’s B).

If you can follow this high value, advanced maths, you should be able to work out why the OP has undercalculated our likely points haul from the remaining fixtures.

Give it time. It will percolate through soon enough.

[SaintBobby]

47 minutes ago, CB Fry said:

"Chance of drawing" isn't a thing you absolute f**king plank.

 

😂😂😂😂 utterly amazing how 10 of our matches have ended in draws if the chance of drawing “isn’t a thing”. What are the odds on that? 😂😂😂😂

[CB Fry]

14 minutes ago, SaintBobby said:

😂😂😂😂 utterly amazing how 10 of our matches have ended in draws if the chance of drawing “isn’t a thing”. What are the odds on that? 😂😂😂😂

That's your best shot now?

Thank you and good night.

[SaintBobby]

3 minutes ago, CB Fry said:

That's your best shot now?

Thank you and good night.

😂😂😂😂

Trying to imagine someone attempting a pre match discussion with CB Fry.

“What do you think our chances of winning are today, CB?”

”Very good. I’d say over 50%”

”Really, I think we will only get a draw, CB”

”I agree. I also think we’ll draw. That is intrinsic to the model”

😂😂😂

[Dark Munster]

Ok, as a mathematician let me step in here. What's missing in the OP's post is the odds of drawing (unless the assumption was that it's 0% for all games, which is, err, wrong).

So suppose over 10 matches against the big boys we have 10% of winning each match, and let's say (for example) 20% chance of drawing, then our expected points from each match is 10% of 3 points + 20% of 1 point + 70% of 0 points = 0.3 + 0.2 + 0 = 0.5 points. So over those 10 matches we'd predict 0.5 X 10 = 5 points.

<Ducks for cover>

[OttawaSaint]

2 hours ago, SaintBobby said:

If you have a really tough run of ten games, where you have a 10% “chance of winning” in each case, how many points would you guess we’d take from those games?

(A) 3 points?

Or (B) More than 3 points?

(I’ll give you a clue. It’s B).

.

I'll be the first to admit that I'm shit at maths but can you explain this? 

If we have 10 game and each has a 10% chance of a win, that means that it was a 90% chance of drawing or losing.

So for that, how can you calculate the number of points we'd take (as being over 3pts)? For me, I don't feel that there's enough information as to what proportion of those 90% are likely to be draws and what are proportion are likely to be losses.

Again, I'm totally shit at maths so what is the explanation for the answer you gave of over 3pts?

Edited 24 January, 2022 by OttawaSaint