Goal Difference!

[Archie's penalty]

I'm convinced Snowball was the editor of Pravda in a previous life.

Just substitute tractor production figures for Ian Hartes' pace and you have Comrade Snowball's glorious Five Game Plan. The figures cannot be questioned. It's 5 years in the Statistical Gulag for you, Comrade.

POTMFY GTFI.

[Snowball]

As a matter of correctness, on "True tables" (ppg) we are eighth according to Statto.

As for allegations that the team is inconsistent, Statto's rating for RFC
has been between 800 and 818 for the last 39 games, with a slow rise
in form until a recent steeper rise. Currently we have equal top rating
with QPR and show a "rising" rating, meaning we might well do better
than the raw stats would predict.

Currently they reckon it could come down to Goal Difference!

1 79 818 QPR
2 74 812 Swansea
3 74 814 Cardiff
4 73 811 Norwich
5 72 808 Watford
6 71 818 Reading
7 71 801 Leeds


Of course, their ratings will not yet take into account signings in the window

No Forest?

They're looking a good bet for 2nd IMO.

My bad, careless cut and paste before I ordered them by predicted points.

They have us as third hottest team and improving. Despite that they only
predict 71 points which is where we should be based on our season so far
and WITHOUT taking into consideration our improved form in the last ten games.

That's based, I presume on a lot of tough games to come


1 79 818 QPR
2 74 812 Swansea
3 74 814 Cardiff
4 74 826 Forest
5 73 811 Norwich
6 72 808 Watford

7 71 818 Reading
8 71 801 Leeds

This seems incredibly tight. Also expect Leicester to challenge, Millwall and Burnley too, and maybe Hull. That's a dozen clubs at least all could genuinely argue they are in with a real chance of the play-offs. We could play solidly for the rest of the season, be a genuinely good side and be 12th!! Crazy league!

I'm hoping that we maintain our improved record, and can get through
Hull, Cardiff, QPR, Norwich, Watford, Millwall with at least 8 points (one win and 5 draws)
and win at Sheffield (11 from 7)... That would have us on 51 from 33 and needing 22-24 points
from our final 13 games (1.7 to 1.8 ppg) to be fairly certain of the POs

[Mr Angry]

I think you are being very optimistic there with your Hull, "Cardiff, QPR, Norwich, Watford, Millwall with at least 8 points (one win and 5 draws)" prediction.

I'd love that, but with away games at Cardiff and Norwich, a defeat in at least one of those 2 is a distinct possibility.

[Snowball]

I think you are being very optimistic there with your Hull, "Cardiff, QPR, Norwich, Watford, Millwall with at least 8 points (one win and 5 draws)" prediction.

I'd love that, but with away games at Cardiff and Norwich, a defeat in at least one of those 2 is a distinct possibility.

Of course. I mean I'd be happy beating Hull and getting 5 draws from the others and beating Sheffield which totals 11 points from 7

But 4 wins and 3 defeats would be 12 points.

Only thing now is not to lose too many of the six pointers, don't let sides create a gap (was all I was thinking)

We have Hull, QPR, Watford and Millwall at the MadStad. Important we don't lose any of those IMO, and if we can nick something at Cardiff or Norwich that will be a big bonus

[Rev Algenon Stickleback H]

Some logic errors in their Arch

Obviously it is generally true that the more goals you score the better you do (EXCEPT...)
Obviously it is generally true that the less goals you concede the better you do (EXCEPT...)

I've posted in the past that both of these have poorer correlations with final position than GD

But really that's not surprising. Some teams score loads of goals but concede loads, maybe even more than they score.
Some teams are very defensive, can shut out the opposition but don't win enough because they don't score enough

That's why GD is a better predictor.

I take it this is something you've proven to be true, with a decent sample size, rather than something you are assuming to be true because it seem logical?

[Rev Algenon Stickleback H]

I PRESENT MY BELIEFS

BASED ON

the statistical evidence.

That's your interpretation.



I base mine on watching the team, seeing improvements (or not) and ALSO using stats to overcome my personal biases.

You don't. You have a habit of making assumptions and basing predictions on your assumptions, without ever checking those assumptions.

A good example been saying players would be better not letting the ball go for a corner as playing on would probably be more productive. You took the surprisingly low conversion rate for corners and made an assumption playing on would be more productive.

Earlier in this thread you countered claims of poor performance by saying that we'd been unlucky in games and had players out of form, as if that's something that only happens to Reading.

Your stats are fine. Your personal bias just leads you to very poor conclusions.

[Snowball]

That's why GD is a better predictor.

I take it this is something you've proven to be true, with a decent sample size, rather than something you are assuming to be true because it seem logical?

Depends what you mean by proof.

I empirically tested my intuition for the CCC and Prem back about ten years

I posted on this separately 1-2 months ago and looked at correlations for goals-scored, goals conceded and GD

It was when we appeared to "not be scoring enough" and I was arguing way back that there are three basic policies

The Real Madrid way: Don't care about goals conceded, just score one more than the opposition.

The Boring-Boring (old) Arsenal way. Rock solid defence, 0-0 or win 1-0

or the balance between.

We saw it with our club this season. We scored OK (5) but conceded (5) in our opening 4 games, swapped the attacking guile of Gylfi for a much more robust defence and then got VERY stingy in defence but appeared to be lacking goal-power. Either there was then a four-goal aberation or McDermot tinkered and suddenly we were scoring 4-4-1-1-3 but conceding too. The HNA "experts" argued that by changing the way we played (in order to get goals) we became vulnerable. We shut up shop again for a while and then brought in Elwood and now we are stingy in defence AND scoring a lot more goals.

5-5 in 4 Games (1.25 For 1.25 against. GD of 0)

Gylfi goes, defence improved

9-5 in 8 Games (1.11 For .63 against) GD of 0.58) (this was distorted by a single 1-3 defeat at Boro, otherwise 8-2 in 7 = 1.14 For .28 against)

worries about scoring, so tinkering...

14-11 in 6 Games (2.33 For 1.9 against) (GD of .43)

and then we buy Elwood, stiffen defensive midfield, allow Hunt & Long to blossom

12-5 in 9 Games 1.33 For .55 against (GD of .78 per game) (includes game without Elwood)
12-4 in 8 Games 1.50 For .50 against (GD of 1.0 per game) (Only "Elwood" games)

[From Despair To Where?]

I'm convinced Snowball was the editor of Pravda in a previous life.

Just substitute tractor production figures for Ian Hartes' pace and you have Comrade Snowball's glorious Five Game Plan. The figures cannot be questioned. It's 5 years in the Statistical Gulag for you, Comrade.

POTMFY GTFI.

And bang on cue, he follows it up with a post laiden with pointless stats that misses the point and demonstrates perfectly why many on here find him so self regarding, humourless and tiresome.

[weybridgewanderer]

7 71 818 Reading
I'm hoping that we maintain our improved record...

So, they agree with me, 7th and just miss out on the play offs

You hope we maintain our improved record

I believe we will continue to struggle against top 6 sides and as a result, over the next 6 weeks, our ppg (and potentially goal difference) will fall from where it is at the moment.

I too "hope" that we maintain our improved record.

I agree that "IF" we maintain our current form we will finish closer to third than 7th, but I don't believe we will.

What do you believe? You continually point out that you have sai d"IF". Also you still haven't offered me a lot of money for a top 5 finish.

I have to believe that you don't really believe it either.

We might scrape 6th, which from play offs perspective would be great. 6th place often outperforms 3rd in the play offs

[Arch]

All about February, for me. If we're in the same condition at the end of that month as we are now, I'd be very confident of ending the year in the top six.

[RoyalJames101]

would be second but for those 5 draws where we could have won most of them with luck or strikers on top form

But I thought you said Shane was on top form?

[Snowball]

Looking at the last nine full seasons in the CCC, these are the GDs that filled the top six places

The top three GDs have always made the play-offs.

Out of 54 places 45 (84%) were a top-six GD, 6 (11%) were 7th-best GD, 1 (2%) was 8th, and just two "freak results"

That is 95% of the top six finishers came from the top 7 GDs, 97% from the top 8

Where teams have been in the top six GDs and failed it's usually very close, often just a goal,
in Cardiff's case they were joint on points, and GD, and missed out on goals scored by a single goal

GDs 1-2-3-4-5-6
GDs 1-2-3-4-5-6
GDs 1-2-3-4-5-7
GDs 1-2-3-4-6-7
GDs 1-2-3-4-7-8.........................GD 5 & 6 came 7th and 8th
GDs 1-2-3-4-6- + Outsider (a crazy year)
GDs 1-2-3-4-7 + (Bristol +1)
GDs 1-2-3-5-6-7
GDs 1-2-3-5-6-7

GD Place and times making the POs

1 100% (9/9)
2 100% (9/9)
3 100% (9/9)

4 78% (7/9)
5 33% (3/9)
6 67% (6/9)
7 55% (5/9)
8 11% (1/9)

Plus two rank outsiders

[weybridgewanderer]

I am sure team with the 3rd best goal difference at the end of the season has, more often than not, made the play offs

Now some important questions

how often has the team in 7th in mid january had the 3rd best goal difference
how often has the team with the 3rd best goal difference in mid january finished the season with the 3rd best goal difference?

[Snowball]

I am sure team with the 3rd best goal difference at the end of the season has, more often than not, made the play offs

EVERY TIME IN THE LAST NINE SEASONS

Now some important questions

how often has the team in 7th in mid january had the 3rd best goal difference
how often has the team with the 3rd best goal difference in mid january finished the season with the 3rd best goal difference?

dunno

and

dunno


But I was REALLY shocked to see how RARE teams coming out of the pack actually are.

If you look at tables after 23 games and after 46 games they are very similar. My subjective opinion
was that there is "always" a team, often two teams who come from nowhere and steal a play-off place.

Not so it seems.

There have been a few cases of teams blowing it and dropping down the table, but not many (if any) cases of 16th to 6th.

[weybridgewanderer]

If you look at tables after 23 games and after 46 games they are very similar. My subjective opinion
was that there is "always" a team, often two teams who come from nowhere and steal a play-off place.

Not so it seems.

There have been a few cases of teams blowing it and dropping down the table, but not many (if any) cases of 16th to 6th.

So we agree then, despite the .91 correlation that at the end of the season the team with the 3rd highest goal difference finishes 3rd, statistically we are unlikely to finish in the play off places, on the whole the league stays pretty much as it is

[weybridgewanderer]

I am sure team with the 3rd best goal difference at the end of the season has, more often than not, made the play offs

EVERY TIME IN THE LAST NINE SEASONS I never disputed this!

Now some important questions

how often has the team in 7th in mid january had the 3rd best goal difference
how often has the team with the 3rd best goal difference in mid january finished the season with the 3rd best goal difference?

dunno

and

dunno

REALLY KIND OF BLOWS YOUR CASE OUT THE WATER!

But I was REALLY shocked to see how RARE teams coming out of the pack actually are.

If you look at tables after 23 games and after 46 games they are very similar. My subjective opinion
was that there is "always" a team, often two teams who come from nowhere and steal a play-off place.

Not so it seems.

There have been a few cases of teams blowing it and dropping down the table, but not many (if any) cases of 16th to 6th.

[weybridgewanderer]

if you had said to us that, over the last 10 seasons, on 7 occasions, at christmas, the team with the 3rd best goal difference has been outside the top six but they usually ended up 3rd or 4th then you had some worthwhile stats

Instead of have picked a stat has a .91 correlation at the end of the season, said we meet that stat now and with no modelling basis behind it extrapolated an outcome.

dogs have 4 legs , cats have 4 legs therefore cats are dogs kind of thing

As another posted might say ... DO YOU SEE ???

[Snowball]

If you look at tables after 23 games and after 46 games they are very similar. My subjective opinion
was that there is "always" a team, often two teams who come from nowhere and steal a play-off place.

Not so it seems.

There have been a few cases of teams blowing it and dropping down the table, but not many (if any) cases of 16th to 6th.

So we agree then, despite the .91 correlation that at the end of the season the team with the 3rd highest goal difference finishes 3rd, statistically we are unlikely to finish in the play off places, on the whole the league stays pretty much as it is

How the hell do you get THAT from what I posted?

I'm saying we very rarely get clubs rocketing from nowhere into the POs. 7th to 6th is hardly "rocketing". 7th to FIRST might be.

Incidentally the .91 correlation does NOT mean that the team with third-best GD is 91% likely to finish third, or best/first etc. The correlation is across ALL teams and shows the general predictability.

In fact at the top end and bottom end the predictability is very strong. In the middle, say 8/9th to 17/18th it's far more mushy

[Snowball]

if you had said to us that, over the last 10 seasons, on 7 occasions, at christmas, the team with the 3rd best goal difference has been outside the top six but they usually ended up 3rd or 4th then you had some worthwhile stats

Instead of have picked a stat has a .91 correlation at the end of the season, said we meet that stat now and with no modelling basis behind it extrapolated an outcome.

dogs have 4 legs , cats have 4 legs therefore cats are dogs kind of thing

As another posted might say ... DO YOU SEE ???

Jeez, mate.

I-WAS-NOT-TRYING-TO-PREDICT-OUR-FINAL-POSITION.

I was saying that IMO we were lower than we should be, based on our VERY good GD
and that augurs well for the final result, since the side with the third-best GD is usually 2-3-4th

Can you not get that?

[weybridgewanderer]

I'll happily agree that if we have the 3rd best goal difference at the end of the season we will make the play offs. NOt coz we have it now, coz we have it at the end of the season. Basically we will have won more games and have gained more points that others around us. Winning games = +ve goal difference

so are you saying you think we will finish 2nd 3rd or 4th or not?

sometimes you say probably, sometimes you say if , sometimes you say if means probably but its by no means certain

come on, what do you predict, top 6 finish? or are you going to stay on the fence and then come back in the summer and say "told you so" no matter the outcome.

I think we will just miss out.

What do you think? What are you trying to say, other than that if we have the 3rd best goal difference at the end of the season we will finish near the top of the table?