Statistical analysis

[Ozyhibby]

One for those interested in stats.
https://www.modernfitba.com/blogs/2018/6/14/rate-my-transfer-florian-kamberi-makes-permanent-switch-to-hibs



Sent from my iPhone using Tapatalk

[CMurdoch]

One for those interested in stats.
https://www.modernfitba.com/blogs/2018/6/14/rate-my-transfer-florian-kamberi-makes-permanent-switch-to-hibs



Sent from my iPhone using Tapatalk

What the fek :confused:
Flawed logic?
Scoring rates depends on
how good the team is that you are playing in,
how good opposing teams defend,
what type of tactics your team and opposition use,
whether you have a strike partner and if so how good that strike partner is.
IMO those and numerous other variables will have a greater baring on a strikers scoring rate than those report stats.

[Elephant Stone]

On the surface, the permanent signing of Florian Kamberi was a no-brainer for Hibs. He came in & scored seven goals in half-a-season & averaged 0.50 goals per 90 minutes (note: I don't include penalties when evaluating players & Kamberi slotted home two of those). That's an average of a goal scored in every other match.

So, 9 goals in 14 to those of us who do count all the goals.

[GlazedDonut]

What the fek :confused:
Flawed logic?

Yeah, I'm a bit confused. I have had a couple of beers tho....

He seems to be saying that players that score a high percentage of shots (which are on target) will score a lower percentage of shots (on target) next season because they will move towards the average percentage for players in the league.

So by that logic you would want to sign a player which hits the target a lot but doesn't score many because they're percentage is bound to improve because of an average of all players? Doesn't sound right be me but i may be misunderstanding.

Aren't some players just better than others?

[SChibs]

On the surface, the permanent signing of Florian Kamberi was a no-brainer for Hibs. He came in & scored seven goals in half-a-season & averaged 0.50 goals per 90 minutes (note: I don't include penalties when evaluating players & Kamberi slotted home two of those). That's an average of a goal scored in every other match.

So, 9 goals in 14 to those of us who do count all the goals.

Makes sense not to include penalties in this analysis.

[Elephant Stone]

Makes sense not to include penalties in this analysis.

I didn't read it past that point so I'm not sure how much sense it makes in the context of that article; it just is a ridiculous piece of writing to say he scored 7 goals for Hibs and had a goal ratio of 1 in 2. That's just not true. He scored 9 goals. Either way, I don't see how it can make sense to ignore those penalties if you're analysing goals per minute and shot accuracy and what have you. He scored two thunderbolt penalties that looked pretty inch perfect to me - I would consider that relevant.

[SChibs]

I didn't read it past that point so I'm not sure how much sense it makes in the context of that article; it just is a ridiculous piece of writing to say he scored 7 goals for Hibs and had a goal ratio of 1 in 2. That's just not true. He scored 9 goals. Either way, I don't see how it can make sense to ignore those penalties if you're analysing goals per minute and shot accuracy and what have you. He scored two thunderbolt penalties that looked pretty inch perfect to me - I would consider that relevant.

If you don't see how it fits in then read the article and it might make more sense. You say it's ridiculous writing but don't read on where the writer, imo, justifies what he says.

[CMurdoch]

Yeah, I'm a bit confused. I have had a couple of beers tho....

He seems to be saying that players that score a high percentage of shots (which are on target) will score a lower percentage of shots (on target) next season because they will move towards the average percentage for players in the league.

So by that logic you would want to sign a player which hits the target a lot but doesn't score many because they're percentage is bound to improve because of an average of all players? Doesn't sound right be me but i may be misunderstanding.

Aren't some players just better than others?

It's Cathro isn't it.
He's back with his laptop again :na na:

[Elephant Stone]

If you don't see how it fits in then read the article and it might make more sense. You say it's ridiculous writing but don't read on where the writer, imo, justifies what he says.

Skimmed through it and still think he's talking out his arse for the reasons I gave above.

[wookie70]

I think it is an interesting read. Players get signed on the basis of a hot spell of form and then never do it again. He is saying that despite one of Kamberi's stats being above the average he is below elsewhere so chance are he will still score goals even if one of teh stats drop as it inevitibly will with most players. I can't claim to understand it all but I hope our scouts use all of these and the more old fashioned eye for a player when choosing our new recruits.

There was one stats that was blindingly obvious when McLaren and Kamberi signed last year. If you get men in the box you will score more goals.

[Hibee Mac]

I do like a bit of statistics, but sometimes people read into things just a bit too much.

I don't understand the guys logic that Kamberi can score so well in the second half of a season but then not keep that up?

[Ozyhibby]

I didn't read it past that point so I'm not sure how much sense it makes in the context of that article; it just is a ridiculous piece of writing to say he scored 7 goals for Hibs and had a goal ratio of 1 in 2. That's just not true. He scored 9 goals. Either way, I don't see how it can make sense to ignore those penalties if you're analysing goals per minute and shot accuracy and what have you. He scored two thunderbolt penalties that looked pretty inch perfect to me - I would consider that relevant.

It’s an analysis of his game in open play. Not sure why that’s so hard to understand?
If McLaren hadn’t missed up in Aberdeen then Flo would not have taken those pens and would have only been on 7 goals. Would that have made him worth less to us? No chance.


Sent from my iPhone using Tapatalk

[Godsahibby]

I read something of his before about why Hearts signed McLean and not play one of the youngsters. His theory McLean was a bad signing which i think we all know but linked to an analysis of young strikers in Scotland. Covered it in some detail but in the end he felt Oli Shaw is the best tallent.

[calumhibee1]

I stopped reading when he said that Maclaren got much more involved in linking up play. Guys never seen Kamberi or Hibs play recently.

[calumhibee1]

I didn't read it past that point so I'm not sure how much sense it makes in the context of that article; it just is a ridiculous piece of writing to say he scored 7 goals for Hibs and had a goal ratio of 1 in 2. That's just not true. He scored 9 goals. Either way, I don't see how it can make sense to ignore those penalties if you're analysing goals per minute and shot accuracy and what have you. He scored two thunderbolt penalties that looked pretty inch perfect to me - I would consider that relevant.

Agreed. I could understand not counting the penalties if he wasn't going to take them next season. But I'd expect Flo to start next season as penalty taker, giving him the opportunity to score penalties next season aswell. Discounting any form of goal is just stupid. Why don't you just take out errors from the opposition, open goals etc aswell and only leave in goals that were from very good finishes or individual pieces of brilliance?

[Kato]

It's Cathro isn't it.
He's back with his laptop again :na na:

Hooray!!

[Vault Boy]

All it really says is that Flo's conversion rate is unsustainable over an entire season. That's probably correct. He's also saying that Flo's shot accuracy is bound to improve as that's also unsustainable, which is also likely to be correct. I wouldn't take these stats to heart, they're purely analytic and obviously don't account for unquantifiable variables such as confidence, potential or partnerships etc. Flo's charts make for good reading.

[Elephant Stone]

It’s an analysis of his game in open play. Not sure why that’s so hard to understand?
If McLaren hadn’t missed up in Aberdeen then Flo would not have taken those pens and would have only been on 7 goals. Would that have made him worth less to us? No chance.


Sent from my iPhone using Tapatalk

He's conducted a pretty thorough statistical analysis of his shot accuracy and goals per minute while ignoring two goals. That, to me, is ridiculous. He scored two high pressure penalties (against Hearts and Rangers) by blasting them into the top corner. That tells you something about his temperament and also his shot accuracy. Likewise, if he had a 100% penalty miss rate I would consider that to be relevant if you're gonna analyse statistics in this amount of detail.

[Hibee Mac]

All it really says is that Flo's conversion rate is unsustainable over an entire season. That's probably correct. He's also saying that Flo's shot accuracy is bound to improve as that's also unsustainable, which is also likely to be correct. I wouldn't take these stats to heart, they're purely analytic and obviously don't account for unquantifiable variables such as confidence, potential or partnerships etc. Flo's charts make for good reading.

This is not a jab at you, more at the guy who writes the article.

Why is Flo's conversion rate unsustainable over a whole season? He did it over a full half season just fine?...

I could understand if Flo had played 2 games all season and scored 1 goal then yes that's not a large enough sample size. But we're talking about half a season here not a couple of games?...

[HFCdeb]

All of this ignores his greatest attributes (in my opinion): the ball sticks to him up top which brings others into play and he bullies defenders. Without him, our team wouldn't have scored as many goals as we did in the second part of the season.
If Kamberi only scores 9 goals the whole of next season, he'll still have added real quality to the way we play.
Grant Holt is one of my favourite ever Hibs players and he didn't score much, his goals tally was irrelevant because of the other strengths he brought to the team.

[Billy Whizz]

This is not a jab at you, more at the guy who writes the article.

Why is Flo's conversion rate unsustainable over a whole season? He did it over a full half season just fine?...

I could understand if Flo had played 2 games all season and scored 1 goal then yes that's not a large enough sample size. But we're talking about half a season here not a couple of games?...

For me Flo isn’t and won’t be our top scorer, it’s what he does that creates goals for others that’s important, although he’ll chip in with quite a few.
That’s why we generally play an out and out goal scorer beside him, like Shaw or Maclaren

[Jim44]

Paralysis by analysis..

[IGRIGI]

What a way to take all the fun out of football.

[Vault Boy]

This is not a jab at you, more at the guy who writes the article.

Why is Flo's conversion rate unsustainable over a whole season? He did it over a full half season just fine?...

I could understand if Flo had played 2 games all season and scored 1 goal then yes that's not a large enough sample size. But we're talking about half a season here not a couple of games?...

Basically just averages. It's certainly not impossible that Flo will continue to convert at the same rate for us next season, and if he does he'll probably be the top goal scorer in the league by a decent way. It's just stastically more likely given his expected goals, shot accuracy and conversion rate, that the figures will balance out over the course of the entire season.

Basically he was on red hot form for us last year. He scored more goals than would have been expected from his analytics. That might continue, or, it might balance out over the course of 40 odd games.

Stats are stats, a poker hand that has 90% equity on the flop is still gonna lose 1/10 times, but it isn't possible to predict when that might be. Pinch of salt with all this stuff! ☺

[A Hi-Bee]

As the heading above says, all figures can be manipulated to bend the truth, just depends on who is doing them.
Lies, Damn Lies and Statistics, what does it all mean.

[Hibee Mac]

Basically just averages. It's certainly not impossible that Flo will continue to convert at the same rate for us next season, and if he does he'll probably be the top goal scorer in the league by a decent way. It's just stastically more likely given his expected goals, shot accuracy and conversion rate, that the figures will balance out over the course of the entire season.

Basically he was on red hot form for us last year. He scored more goals than would have been expected from his analytics. That might continue, or, it might balance out over the course of 40 odd games.

Stats are stats, a poker hand that has 90% equity on the flop is still gonna lose 1/10 times, but it isn't possible to predict when that might be. Pinch of salt with all this stuff! ☺

Cheers for the explanation, makes a bit more sense 👍

[Eaststandee]

I stopped reading when I seen "written by: @TheGersReport." :rolleyes:

[ballengeich]

The writer asserts that players who have done something better (or worse) than average for a while are likely to move closer to an average performance over a longer period. That assumes that achievement in football is a random matter. What he needs to take into account is that players sometimes achieve more (less) than average because they're good (useless).

[KWJ]

I think he is saying they were below average on their career last year so expect to see a bounce back. See a lot more of this in american sports and theres some truth to it.

[ballengeich]

I think he is saying they were below average on their career last year so expect to see a bounce back. See a lot more of this in american sports and theres some truth to it.

He's comparing stats with the league average rather than Kambieri's career average. Even if he was using a player's career average I'd want to look at the stage of his career the player had reached.

I'm not dismissing statistical analysis of performance. I'm just not convinced by the approach this article's taking.

[The Leith Dutch]

That's a good analysis there.
Some very interesting stuff.

Not sure what the fuss is about him not including penalties - it's pretty clear from what he says that he's interested in the underlying statistical basis of how many goals a player is likely to score and penalties are, statistically, a completely different animal. The most significant factor in predicting the likely return would be whether you were the designated taker or not.

He's not discounting their value as goals.

[Treadstone]

I like stats up to a point. Where I think the article is unfair is the assist part of it. Kamberi assisted no goals in the 5-5 Rangers game but anyone watching would realise how much he worked the Rangers defence created space for others and brought teammates into the game. No stat for that but that's why scouts are employed as well as analysts. Stats say Mcclaren 3 goals Kamberi 1. I'm grateful for Jamies goals but not possible without Kamberis tireless shift.

[Ozyhibby]

I like stats up to a point. Where I think the article is unfair is the assist part of it. Kamberi assisted no goals in the 5-5 Rangers game but anyone watching would realise how much he worked the Rangers defence created space for others and brought teammates into the game. No stat for that but that's why scouts are employed as well as analysts. Stats say Mcclaren 3 goals Kamberi 1. I'm grateful for Jamies goals but not possible without Kamberis tireless shift.

That’s as stats should be read. They are an aid to decision making which helps scouts minimise their own bias.


Sent from my iPhone using Tapatalk

[1van Sprou7e]

How on earth has he come to the conclusion that Maclaren was more of a playmaker than Kamberi

Maybe if he actually watched a game rather than basing everything off of stats...