Analysis of seasons past- results surprise

[jirskyr]

By the way i think you hve just provided enough ammo for JT's next 8 Press conferences.

They should be called - 101 different ways say were improving despite coming last

Well depending on what stats you choose to look at, as you said, yes you can interpret in many ways.

It's definitely true that we are an improved defensive outfit in 2015, compared to the 15 former versions of ourselves. What I really wanted to understand is why our win record is so bad, if the defence isn't really so bad.

No matter what Taylor says about improvements, it's total wins that matter the most, and in that regard we are not making progress.

 

[jirskyr]

Do you think Wayne Bennett rattles on about stats? I reckon he just knows how to man mange each individual and create a environment where everyone wants to run through a brick wall for one another.

I actually suspect Bennett and most other coaches do rattle on about stats, a lot. Or if not them personally, they employ someone to do it. I know Des Hasler is mad on this, especially the performance analysis of his players.

Not that the best coaches don't also work on a winning and committed culture. But all professional clubs are highly analytical about their preparation and review, whether it be on-field or fitness management. That is how professional teams work nowadays.

 

[Russta]

86% of statistics are made up. :laughing:

 

[jirskyr]

That said - I wouldn't mind seeing a rolling average of our average points against. My concern is that this trend is going the wrong way, after a pretty decent start.

It's actually the opposite - recently we tend to lose heavily in the first game and then see a substantial early improvement in average points conceded.

Interestingly 2015 started off way better than 2013-2014, but is heading back in towards them.

There is a weakness in this way of looking at the data however. A rolling average, by definition, means results will start to plateau as more and more datapoints are included. We would always expect early outliers because they are the average of only 1 or 2 games, and these games could be blow outs. But by the time you hit Round 26, one single match is only contributing to 1/26th of the average.

 

[Eddie]

First 8 weeks of the comp - Under dogs have their chance as they have had a longer pre season, more to prove. Bigger sides are under done or carrying a finals hangover. All teams are rusty so the element of randomness in a game goes up further.

After first 8 weeks - The grind kicks in and the cream rises to the top. Teams start to play more consistently (good or bad) The weak rosters and mentally weak sides start to get blown pout and becomes all too hard.

 

[Mickeyboy]

Rolling average means we are heading over the 2013 and 2014 results, look for the spike after Friday night :bawling

 

[ricksen]

That said - I wouldn't mind seeing a rolling average of our average points against. My concern is that this trend is going the wrong way, after a pretty decent start.

It's actually the opposite - recently we tend to lose heavily in the first game and then see a substantial early improvement in average points conceded.

Interestingly 2015 started off way better than 2013-2014, but is heading back in towards them.

There is a weakness in this way of looking at the data however. A rolling average, by definition, means results will start to plateau as more and more datapoints are included. We would always expect early outliers because they are the average of only 1 or 2 games, and these games could be blow outs. But by the time you hit Round 26, one single match is only contributing to 1/26th of the average.

Thanks.
I actually meant the trend specifically in 2015, but good to have the reference points of previous years.

You're right about the rolling average being a weak metric, but the trend is there - both in the numbers and the eye-test.
I was pretty impressed with our defensive attitude over the first several rounds as it was in such stark contrast to years previous. For whatever reason this attitude hasn't been seen in months, which is hugely concerning to me. Some of the crap we saw on Sunday (first half at least) was just as bad as anything we saw in the post Tallis-gate era late last season.
I'd hate to think JT was already losing his grip on the dressing room already.

 

[phil1986]

I appreciate a post like this. It is purely analytical, not emotions involved.

Unlike some comments that have and will follow it.

_Posted using RoarFEED 4.2.0_

 

[Tigerdave]

excellent analysis jirskyr, very interesting stuff. It did appear we have been going alright defensively as we haven't been smacked too many times so far this year.
\

While I can understand the reason for a comparison , I went through in a previous post , the number of variables that are present in such a comparison,
Just a few were: did we play more or less night games.
We're there a lot of wet weather games.
What injuries did we have in comparison, or what injuries did the opposition have
were the opposition better or worse in those years
Were rep players out Of our team or theirs
What suspensions did we or the opposition have
That's just a few things that can and do effect how a team plays on any specific day
It's not scientific , but shows how hard it is to compare even two seasons and be any where accurate in any findings
I'm not knocking your comparison.
Just pointing out the difficulty in getting answers

yeah lots of other factors to consider.

I think it would be a forgone conclusion we've had less injuries this year than in others, certainly in the bad years. Wouldn't be surprised if we've had less than 2005, 2010 and 2011 as well though.

As for rep teams, we'd only need to worry about State of Origin though right? When did they start implementing the standalone weekend for the Anzac Day test match and Country/City Origins?

Haven't missed too many for Origin, Robbie and now Woods. Sauce, Hodgo, Prince and Terry Hill…...........that's been it.

 

[NT_Tiger]

Thanks for the effort and insight jirskyr.

 

[Tiger_In_The_Gong]

That said - I wouldn't mind seeing a rolling average of our average points against. My concern is that this trend is going the wrong way, after a pretty decent start.

It's actually the opposite - recently we tend to lose heavily in the first game and then see a substantial early improvement in average points conceded.

Interestingly 2015 started off way better than 2013-2014, but is heading back in towards them.

There is a weakness in this way of looking at the data however. A rolling average, by definition, means results will start to plateau as more and more datapoints are included. We would always expect early outliers because they are the average of only 1 or 2 games, and these games could be blow outs. But by the time you hit Round 26, one single match is only contributing to 1/26th of the average.

As we can see by the chart at round 17 2015 we are almost at the same level we were at round 17 2014.

Now add round 18 where we are beaten 38-6.

Then round 19 where we are beaten 42-16

My money is on Jason Taylors line being above Mick Potters line after the 19 rounds played.

Anyone getting excited about these stats and not taking the last 2 rounds into consideration are deluding themsleves.

 

[Snake]

Oh dear Taylor will see these stats and Mr Motor mouth with have verbal diareaha…. Question do stats allow you to play in finals footy ,do they put bums on seats ,do they make a supporter spend there hard earned to sign up for memberships,do they give sponsors the encouragement to sign on ,do they inspire players to sign.
When the team that you follow is last in the competition stats are worthless and the only place for them is in the shredder ,winners do not read stats there more interested in playing FINALS football and surrounding themselves with positive people not sitting alone in the coaches office reading stats!

 

[Tiger_In_The_Gong]

I have added rounds 18 and 19 to the original data.

Our attack is now averaging 17.42 points a game - **The worst ever by a Wests Tigers team**.

Our defense is now averaging 24.36 points a game - **Only the 10th best for a Wests Tigers team**
\
\
So lets get this clear and cut all the spin -

***16th out of 16 in attack\
\
*10th out of 16 in defense.**

Is it any wonder we are coming dead last.

 

[guyofthetiger]

Mirrors somewhat Wally's comments as per other recent thread.

Just a few moments in a game where we make key errors, but otherwise we're not too far off.

What's to be expected as the team get some games under their belt.

As for the spoon, well, that was always on the cards this this year.

It was more a surprise that for the briefest minute it looked like we could miraculously do ok this year.

 

[supercoach]

Be very interesting to see how the stats stack up at the end of the season. With the season still incomplete it makes it very hard to read to much into the stats.

Also very hard to compare seasons,so many variables,injuries,the draw,rep requirements and so on,but the bottom line their are 15 teams winning more games than us and that is the object of the game…....TO WIN GAMES

 

[stevetiger]

I have added rounds 18 and 19 to the original data.

Our attack is now averaging 17.42 points a game - **The worst ever by a Wests Tigers team**.

Our defense is now averaging 24.36 points a game - **Only the 10th best for a Wests Tigers team**
\
\
So lets get this clear and cut all the spin -

***16th out of 16 in attack\
\
*10th out of 16 in defense.**

Is it any wonder we are coming dead last.

I think at the moment we are just a poorly coached side with some deficiencies in our squad.

There is no way to spin our place on the ladder and I think its pretty obvious that our defence is ordinary and our attack has gone backwards despite having a bunch of exciting attacking players.

JT deserves to cop it for a terrible performance in his first year coaching us. Lets hope he improves but lets not try and spin it into something that it isn't.

 

[Tiger_In_The_Gong]

I have added rounds 18 and 19 to the original data.

Our attack is now averaging 17.42 points a game - **The worst ever by a Wests Tigers team**.

Our defense is now averaging 24.36 points a game - **Only the 10th best for a Wests Tigers team**
\
\
So lets get this clear and cut all the spin -

***16th out of 16 in attack\
\
*10th out of 16 in defense.**

Is it any wonder we are coming dead last.

I think at the moment we are just a poorly coached side with some deficiencies in our squad.

There is no way to spin our place on the ladder and I think its pretty obvious that our defence is ordinary and our attack has gone backwards despite having a bunch of exciting attacking players.

JT deserves to cop it for a terrible performance in his first year coaching us. Lets hope he improves but lets not try and spin it into something that it isn't.

I agree, these Taylor apologists have looked for any justification that he was doing a good job.

Now we have statistical evidence that as of round 19 he has destroyed our attack and the so called defensive revival has bottomed out and is nowhere near as positive as has been represented.

Not that stats are the be all and end all but its hilarious they use these stats to justify his restructuring of our team yet the exact same stats if presented in a fair and up to date manner show him to be the abysmal failure others have claimed :laughing:

 

[Lakra]

Injuries… This year we've had hardly any and yet we are most likely getting the spoon. Previous years we had at least 1-2 players injuried in some way every single game. Potter did crazy well considering his options and what was happening at the club. Taylor has a full team and cannot win with them..

 

[jirskyr]

I have added rounds 18 and 19 to the original data.

Our attack is now averaging 17.42 points a game - **The worst ever by a Wests Tigers team**.

Our defense is now averaging 24.36 points a game - **Only the 10th best for a Wests Tigers team**
\
\
So lets get this clear and cut all the spin -

***16th out of 16 in attack\
\
*10th out of 16 in defense.**

Is it any wonder we are coming dead last.

No this is incorrect. Everything I have posted in this thread is all results up to and including Round 19 vs Broncos.

We didn't even play Round 18, it was a bye.

Attack in 2015 is 14/16 of all seasons as measured by "Points For per game", defence is 6/16 for Points Against.

We are coming last 2015 for several reasons, and it is not just about points scored or conceded. Firstly, our losing margins are quite small, but they are consistent. I.e. we lose often but not by much = we are unable to convert close games into wins. That is the primary issue with our 2015 season: we either win big, or we lose.

We also happen to be caught in an unfortunately better performing bottom ladder.

For the past 5 seasons 2010-2014, the bottom placed side has totalled 5 or 6 wins (for 2010 Melbourne were moved to last due to salary cap, but 2nd last Cowboys only had 5 wins). We are obviously already on 5 wins with 7 games in hand.

Furthermore, the win % for bottom teams in 2015 is higher than any of the previous 5 seasons:
- bottom 4 teams win % 2010-2014 = 30.63%; in 2015 = 33.82%
- bottom 3 teams win % 2010-2014 = 28.33%; in 2015 = 33.33%
- bottom 2 teams win % 2010-2014 = 26.25%; in 2015 = 32.35%

In other words, the bottom 4 are doing better this year than any of the previous 5 years. Even further to this, the average difference between 8th and last for the previous 5 years was 0.55 table points per game. It is currently 0.47 table points per game, so we are closer to the Top 8 than last usually is.

So it's not just about Tigers doing badly in 2015, it is also that bottom-performing sides this year are doing better than they usually do.

It amuses me that you say "lets get clear and cut out the spin" and then proceed to spin/interpret the data the way you see fit.

Data is data, it is facts. Interpretation of data is always prone to bias; as others have said you may be able to support opposed arguments by selective interpretation of the same data.

But don't make the mistake of calling one argument "spin" and another "clear". All arguments about human behaviour are spin of one kind of another.

 

[jirskyr]

Be very interesting to see how the stats stack up at the end of the season. With the season still incomplete it makes it very hard to read to much into the stats.

Not really. Because the season is incomplete, I made sure to focus on comparison of metrics per match, rather than season total. Not all seasons are of same length anyway.

If you average out the results per game, they are very much comparable.

The only argument you might have is that the final results could be heavily influenced by the final rounds of the competition. I.e. it may be true (I haven't looked into it) that bottom sides perform consistently worse once the season is a write-off (or top sides get even better as they gear up for the finals). We may still salvage some pride for 2015, or we may fall right through the bottom.

But you make an interesting point - if you do believe that it is hard to read much into stats for an incomplete year, how well can you truly criticise / praise a coach and team until the year is over?

Also very hard to compare seasons,so many variables,injuries,the draw,rep requirements and so on,but the bottom line their are 15 teams winning more games than us and that is the object of the game…....TO WIN GAMES

It makes another point - if you say that results are hard to compare between seasons, then you cannot simultaneously say "worst season ever" or "worst coach ever". Either you compare to previous efforts, or you look at each season as its own dataset.