France v Ireland 4.45pm Sat 3rd February 2018

[fourthirtythree]

O'Mahony can't get over the gainline to save his life - 64 carries for 70m for Munster this season [source: https://munsterrugby.ie/player?PlayGuid=PO196534 ] – it's what you'd expect from a prop, not a blindside. It's noticeably lower than Devin Toner's 63 carries for 80 metres [source: https://www.leinsterrugby.ie/player?PlayGuid=DT150994 ], and he's supposed to be the worst carrier in the world.

At 1.91m tall POM would have better yardage going forward if he put the ball above his head and litteraly just fell forward.

I'm not doing the maths but I'm sure Devin would too!*









*Yes I do shout "stretch for the line" every single time Toner is tackled on the halfway line, it's a ritual at this stage.

[Peg Leg]

There seems to be 0 evidence supporting POM starting this one. His carries and tackle count have been seriously low. Anonymous in most games I've seen him in, except for looking pissed off with referees. Especially when you consider how well Leavy and Conan have been playing.
That said, POM seems to always put in very good performances for Ireland, so here's hoping he finds that bit of form.

That bit about his referee management should have been enough to see him dropped IMO.
This is why JSsnr drives me nuts. I can't see the rationale behind some of his selections because the metrics don't add up, but I also have no sense of the team mood music or what are the intangibles that make them tick. It's very much the latter category that i feel Peter falls into and as cliche'd as this sounds he is never found lacking in the passion/effort stakes. Some teams (as i write this i think something similar may have been discussed on BTTV) play better when certain individuals are playing regardless of their physical contribution. I think that POM & Heaslip (not to question his physical impact) are those players for Ireland. In the case where Heaslip is fit I would expect to see Stander wearing 6.

[suisse]

I generally think in rugby and football, we're a little behind the Americans when it comes to stats. Stats are vitally important. I love stats. I think everything has to be given a number and something without a number is essentially unimportant or not important enough to rate. Other key stats like meters made and tackles made etc. I'd be really interested to know where a certain player makes his meters and where most of his tackles are made. Someone on Newstalk said Keith Earls bides his time and essentially rips a team apart in the last 20 mins. He can play at 100% for 80 mins whereas other players drop to 90% etc. I wonder is this backed up with numbers or just a hunch? Does Earls make more yards, score more tries and beat more defenders in the final 20mins? Or, like leadership, is it just something you think you see?

[Peg Leg]

I generally think in rugby and football, we're a little behind the Americans when it comes to stats. Stats are vitally important. I love stats. I think everything has to be given a number and something without a number is essentially unimportant or not important enough to rate. Other key stats like meters made and tackles made etc. I'd be really interested to know where a certain player makes his meters and where most of his tackles are made. Someone on Newstalk said Keith Earls bides his time and essentially rips a team apart in the last 20 mins. He can play at 100% for 80 mins whereas other players drop to 90% etc. I wonder is this backed up with numbers or just a hunch? Does Earls make more yards, score more tries and beat more defenders in the final 20mins? Or, like leadership, is it just something you think you see?

I'm with you on the stats, just listened to a piece on the statistical impact of cultural events.
https://mobile.nytimes.com/2018/01/30/u ... -31-18.htm
But in sport, some of the stats need to be measured at a team level rather than person specific and the sampling needs to be like for like before being used as a referenced measure. Frankly there are too many variables, important to too many performance subsets to provide a measured platform to identify successful patterns.
Its like the whole murder rate : snes sales in detroit (im almost certain it was detroit) line graph matching perfectly. You can present that as a piece of data to demonstrate causation and be totally undermined by the murderer age:snes owner age graph.
EDIT: my point- are we more successful against like for like opposition with POM playing, I would guess- yes [looks over shoulder to check to see if xan is sneaking up behind me to smash me over the head with the FACTS]

[Laighin Break]

I generally think in rugby and football, we're a little behind the Americans when it comes to stats. Stats are vitally important. I love stats. I think everything has to be given a number and something without a number is essentially unimportant or not important enough to rate. Other key stats like meters made and tackles made etc. I'd be really interested to know where a certain player makes his meters and where most of his tackles are made. Someone on Newstalk said Keith Earls bides his time and essentially rips a team apart in the last 20 mins. He can play at 100% for 80 mins whereas other players drop to 90% etc. I wonder is this backed up with numbers or just a hunch? Does Earls make more yards, score more tries and beat more defenders in the final 20mins? Or, like leadership, is it just something you think you see?

I'm with you on the stats, just listened to a piece on the statistical impact of cultural events.
https://mobile.nytimes.com/2018/01/30/u ... -31-18.htm
But in sport, some of the stats need to be measured at a team level rather than person specific and the sampling needs to be like for like before being used as a referenced measure. Frankly there are too many variables, important to too many performance subsets to provide a measured platform to identify successful patterns.
Its like the whole murder rate : snes sales in detroit (im almost certain it was detroit) line graph matching perfectly. You can present that as a piece of data to demonstrate causation and be totally undermined by the murderer age:snes owner age graph.
EDIT: my point- are we more successful against like for like opposition with POM playing, I would guess- yes [looks over shoulder to check to see if xan is sneaking up behind me to smash me over the head with the FACTS]

It was BOD that mentioned that about Earls. Sounded like it was more of a hunch than anything else.

I haven't watched it yet, but supposedly this is a good stats-based analysis on the Six Nations teams:
https://www.youtube.com/watch?v=kd8AEnY ... e=youtu.be

[fourthirtythree]

New levels of delusion and paranoia down in Southern Province judging by the latest post on munsterfans.com

I just had a look and it seems that the post-truth Heaslip truthers are out in force: Joe has made up Conway's injury apparently.

FFS they should have a little more cop on than that given just how magnificently stupid so, so many people were revealed to be in public with their Heaslip nonsense.

[paddyor]

New levels of delusion and paranoia down in Southern Province judging by the latest post on munsterfans.com

I just had a look and it seems that the post-truth Heaslip truthers are out in force: Joe has made up Conway's injury apparently.

FFS they should have a little more cop on than that given just how magnificently stupid so, so many people were revealed to be in public with their Heaslip nonsense.

I enjoyed the part where they made the case for Leaflet over Carbery

[LeRouxIsPHat]

You don't think he'd help us get off to a flyer?

[Peg Leg]

You don't think he'd help us get off to a flyer?

You'll be needing your hat sir?

[Oldschool]

The fundamental thing that most people don't understand about stats is that it's not the numbers that are important it's the interpretation of the numbers that is important.
The expertise required to interpret statistics properly is beyond the wit of any poster, moi included, that posts here.
Joe may have access to this sort of expertise but it wouldn't come cheap unless the individual concerned is a rugby fan and does it for fun.
Asking the right question never mind getting the right answer might and I mean might be within JS' ken but it's a fair bet that he'd need a lot of tutoring.
JS however does have that capacity because he's so thorough he's bound to be aware of the usefulness of properly understood/analysed statistics.

[Oldschool]

So basically lads f^ck the statistics and tell us how we're going to beat France tomorrow.
Better still tell us how JS thinks we're going to beat France tomorrow.

[domhnallj]

The fundamental thing that most people don't understand about stats is that it's not the numbers that are important it's the interpretation of the numbers that is important.
The expertise required to interpret statistics properly is beyond the wit of any poster, moi included, that posts here.
Joe may have access to this sort of expertise but it wouldn't come cheap unless the individual concerned is a rugby fan and does it for fun.
Asking the right question never mind getting the right answer might and I mean might be within JS' ken but it's a fair bet that he'd need a lot of tutoring.
JS however does have that capacity because he's so thorough he's bound to be aware of the usefulness of properly understood/analysed statistics.

Speak for yourself! It’s the question you ask first then the interpretation.

[Raydollard]

Stats show a couple of wins and one draw in France in 40 years. Given those stats which are easy to analyse we should win tomorrow.

[hugonaut]

The expertise required to interpret statistics properly is beyond the wit of any poster, moi included, that posts here.

It's really not. Leinsterfans is a broad church. There are people who post here who are so ITK they pretty much invented the subject [I'm not one of them].

Nobody with a hint of sense looks at statistics [i.e. collected pieces of information] and ignores the visual record of what happened on the pitch, just as nobody with any significant interest in rugby looks at a game without critically appraising it through the media open to them [replays, statistics etc.]

[paddyor]

I generally think in rugby and football, we're a little behind the Americans when it comes to stats. Stats are vitally important. I love stats. I think everything has to be given a number and something without a number is essentially unimportant or not important enough to rate. Other key stats like meters made and tackles made etc. I'd be really interested to know where a certain player makes his meters and where most of his tackles are made. Someone on Newstalk said Keith Earls bides his time and essentially rips a team apart in the last 20 mins. He can play at 100% for 80 mins whereas other players drop to 90% etc. I wonder is this backed up with numbers or just a hunch? Does Earls make more yards, score more tries and beat more defenders in the final 20mins? Or, like leadership, is it just something you think you see?

I dunno. I might have mentioned it already (back in the Adam Byrne tackle stats debacle), but I read a book by Michael Lewis (of Moneyball fame) called the Great Undoing Project about 2 Israeli ssychologists. Opening chapter was about a sports teams use of data. They went a lot further than just cold hard stats and used scouts to get more info like upbringing, work out etc. But eventually (like an interview) candidates learn to game the system, and everyone is singing from the same hymn sheet. They end up having to investigate the scouts bias and tell telling the scout to try and ignore the bias (for more than just personal stuff). It's a rabbit hole to some extent.

FWIW I remember an interview with DOC where he talked about lads tarting up their stats in the dying minutes of a game with 2-3 tackles so maybe the hard stats like tackles are dubious to some extent as well. OTOH beastmode will boost your stats too:

https://www.youtube.com/watch?v=aOqtyHPPEJ0

Tackles:
McGrath 78/12 for the season off 789 mins (10.9 mins per tackle -86% success)
Murray 68/12 for the season off 965 mins (14.1 mins per tackle - 85% success)
Cooney 70/30 for the season off 1097 mins (15.6 mins per tackle - 70% success)
Marmion 41/4 for the season off 708 mins (17.2 mins per tackle - 91% success)

Nobody fakes beastmode over the course of a season. There's one outstanding defender there, and it's clearly not the guy at 91% success. FYI if you check last seasons stats they're similar. Injuries aside (and assuming they have similar game minutes) McGrath will finish the season with a higher tackle completion number(+20). If it wasn't Italy at home next up with a 2 week break after, I reckon he'd get more minutes against France.

PS: not interested in Scrumhalf day.

[paddyor]

It's really not. Leinsterfans is a broad church. There are people who post here who are so ITK they pretty much invented the subject [I'm not one of them].

Nobody with a hint of sense looks at statistics [i.e. collected pieces of information] and ignores the visual record of what happened on the pitch, just as nobody with any significant interest in rugby looks at a game without critically appraising it through the media open to them [replays, statistics etc.]

Typically, the problem is the people compiling the stats.

[neilinboston]

I asked a friend, who was responsible for implementing statistical research at Harvard law, what stats can tell...
He said ‘nothing, without a good question.’
Stats in my hands are pointless, I can find you a million meaningless correlations. But you don’t have to be Nate silver to be able to use them effectively.

[Oldschool]

The expertise required to interpret statistics properly is beyond the wit of any poster, moi included, that posts here.

It's really not. Leinsterfans is a broad church. There are people who post here who are so ITK they pretty much invented the subject [I'm not one of them].

Nobody with a hint of sense looks at statistics [i.e. collected pieces of information] and ignores the visual record of what happened on the pitch, just as nobody with any significant interest in rugby looks at a game without critically appraising it through the media open to them [replays, statistics etc.]

It really is. The understanding and proper use of statistics is well outside the ability of most people.

[Peg Leg]

The expertise required to interpret statistics properly is beyond the wit of any poster, moi included, that posts here.

It's really not. Leinsterfans is a broad church. There are people who post here who are so ITK they pretty much invented the subject [I'm not one of them].

Nobody with a hint of sense looks at statistics [i.e. collected pieces of information] and ignores the visual record of what happened on the pitch, just as nobody with any significant interest in rugby looks at a game without critically appraising it through the media open to them [replays, statistics etc.]

Very much this

[Oldschool]

The expertise required to interpret statistics properly is beyond the wit of any poster, moi included, that posts here.

It's really not. Leinsterfans is a broad church. There are people who post here who are so ITK they pretty much invented the subject [I'm not one of them].

Nobody with a hint of sense looks at statistics [i.e. collected pieces of information] and ignores the visual record of what happened on the pitch, just as nobody with any significant interest in rugby looks at a game without critically appraising it through the media open to them [replays, statistics etc.]

Very much this

I can see this is going to get tedious so I'll leave you to it.