The nutty professor

  • We had an issue with background services between march 10th and 15th or there about. This meant the payment services were not linking to automatic upgrades. If you paid for premium membership and are still seeing ads please let me know and the email you used against PayPal and I cam manually verify and upgrade your account.
  • We have been getting regular requests for users who have been locked out of their accounts because they have changed email adresses over the lifetime of their accounts. Please make sure the email address under your account is your current and correct email address in order to avoid this in the future. You can set your email address at https://silvertails.net/account/account-details
  • Wwe are currently experience some server issues which I am working through and hoping to resolve soon, Please bare with me whilst I work through making some changes and possible intermittent outages.
  • Apologies all our server was runing rogue. I managed to get us back to a point from 2:45 today though there is an attachment issue i will fix shortly. Things should be smooth now though
Interesting, and someone already posted it here last week I think

It does show the detail and thought that goes into Manly's planning,which is impressive.

This bit though - 2017 to 2021 - may as well scrap the stats before they changed the rules to 6 again, because that's such a fundamental change in the game?
Haidar found that in the period between 2017 and 2021, across the NRL, the correlation between “possessions” – the number of times the ball is received in hand- and win percentage was 0.76, a close relationship.
 
This is a really interesting insight into our tactics, on the basis of Haidar’s analysis we adopted an expansive game plan of maximum use of the ball and based on the above analysis that is a sound tactic.

However it’s a flawed ideology. Any statistician worth his salt knows that singular correlations don’t imply causation and there are a number of factors that need to be assessed.

The first flaw is that this is a univariate analysis and doesn’t account for intercorrelation of any other factors, it adopts the simplistic conclusion that the more times you pass the ball the more likely you are to win. This data would likely be skewed because teams playing from in front with momentum would logically play more expansive footy resulting in more passes.

The second flaw is that a multivariate analysis would account for trends far more reliably. In a previous post I pointed out the correlation between the top sides/number of wins to percentage of play the balls in oppositions own half.

I’d really like to see the interaction term calculated between Haidar’s analysis and the territory case above.

I’m summary we developed a game plan based on what I believe is a flawed premise, there’s certainly merit in their conclusions but they must be modified as they resulted in the huge defensive pressure we placed ourselves under.

It goes back to the old anecdotal premise of earning the right to use the ball, if we modified this to a hierarchical system of objective one, win the field position battle and then from this point play expansive footy, by which I mean play as expansively as we want to in good ball areas while prioritising controlling the game and maximising turning over possession as close to the opposition goal line as possible we’d be in far better shape and if we did this we would probably have won the comp this year.
 
Yeah , I dunno about that ****.

To me 64% possession to South’s and 36% to us by half time determined the outcome of the PF match.

It was “ game over “ by then.
You’re right Mark, another flaw with their conclusion is that the more Possession you have would also be highly positively correlated to what the Manly stats team define as “possessions”.

Clearly the Manly stats team have gotten this very wrong, in fact it wouldn’t even pass statistics 101 scrutiny.
 
Disclaimer: If the Manly stats guys are reading I do understand this has been described by Roy Masters who isn’t exactly a Rhodes Scholar and would have bastardised the theories.
 
Disclaimer: If the Manly stats guys are reading I do understand this has been described by Roy Masters who isn’t exactly a Rhodes Scholar and would have bastardised the theories.
Yes. It is possessions sure, but also what you do with it. For example giving the ball to turbo 35 times instead of 20. And with other players running in good shapes.
 
Use the ball and complete your sets will out weigh completing your seats and not using the ball
Eg bulldogs completed well but couldn't do anything with it
Manly used the ball but made too many errors with it but still were one game from a GF
Fairy simple game really , run hard tackle hard and hold on to thing will win you more games than you loose
Sometimes people make things harder than they are
 
Manly love to play an up tempo game and that comes with a higher risk and we got exposed in two of the finals.

What must be learned is to have a more balanced approach. Earn the right to be flamboyant by going set for set and gaining field position and being lethal on the counter attack.

The two finals we lost really highlighted that as we conceded many tries direct from our turnovers. It is something that a young team can learn and we need to be smarter and selective when attacking.

I think a few other options rather than Tommy will give our side a lot more variety as right now the top sides know they can shut us down by not giving Tom and DCE space. They are our two biggest weapons and in 2022 look to Schuster to take on more responsibility and give us a strong option to improve our attack.
 
I think the solution lies in the comparison that was given between the 2019 Premier Roosters and the 2021 Bulldogs ....and that is if you have a roster like the Roosters did ... and a roster like the Bulldogs had ... it doesn't make a damn difference how many times you complete or use the ball .... quality players across the park is far more important ..
 
Manly love to play an up tempo game and that comes with a higher risk and we got exposed in two of the finals.

What must be learned is to have a more balanced approach. Earn the right to be flamboyant by going set for set and gaining field position and being lethal on the counter attack.

The two finals we lost really highlighted that as we conceded many tries direct from our turnovers. It is something that a young team can learn and we need to be smarter and selective when attacking.

I think a few other options rather than Tommy will give our side a lot more variety as right now the top sides know they can shut us down by not giving Tom and DCE space. They are our two biggest weapons and in 2022 look to Schuster to take on more responsibility and give us a strong option to improve our attack.
Yes, and many of our big scoring games had lots of tries in the last 10 min of each half. Showing that we did the proper groundwork.
 
Manly love to play an up tempo game and that comes with a higher risk and we got exposed in two of the finals.

What must be learned is to have a more balanced approach. Earn the right to be flamboyant by going set for set and gaining field position and being lethal on the counter attack.

The two finals we lost really highlighted that as we conceded many tries direct from our turnovers. It is something that a young team can learn and we need to be smarter and selective when attacking.

I think a few other options rather than Tommy will give our side a lot more variety as right now the top sides know they can shut us down by not giving Tom and DCE space. They are our two biggest weapons and in 2022 look to Schuster to take on more responsibility and give us a strong option to improve our attack.
I think we are kind of saying the same thing.

We have more firepower than any team and can score from everywhere on the park and high risk can equal high reward.

However if we make those risks more calculated we will have far better results.

Conversely to your point the number of errors is an important stat. A high number of errors would seemingly be bad but a team with 8 errors made on the attack at the other end of the field is probably at a huge advantage to a team who makes 4 errors bringing it out of their own end.

Like all things higher risk equals higher reward but the risk/return profile changes based on field position. Playing risky footy inside their 20 has a lower element of risk compared to throwing an offload bringing it off our own line.

When we are in our end it’s far more productive to play percentage footy, kick to corners and build defensive pressure instead of trying to overplay our hand.

The closer we get to their try line the lower the cost of an error so it would seem to me that if we win the field position battle then unleash our attacking weapons in areas where errors have lower average cost is a no brainer.

One final comment on possession it also makes sense that in a game that say for example Manly have 40% of possession but win the battle of field position to 60% of tackles in their own half the possession stat would be balanced.

Say the opposition have 24 sets to our 16 but of those sets only 10 are in our half, Manly would also have 10 in their half which would amount to roughly equal time under high defensive pressure.

Focus on dominating field position then unleash fire power in good ball. Errors don’t matter in isolation, what matters is where you make your errors.
 
Yeah this article made me realise where we went wrong against the top teams (Souths/Penrith/Melbourne) and i have posted about it/referenced it in other threads.

And it also explained why we didn't change tactics (to focus on the forward grind/completion of sets etc) despite being shown by losses to the Storm that this didnt work against the top teams.

Quite simply we decided to live by the sword and die by the sword under all circumstances.

As other posters have mentioned hopefully Des can find a better way to balance this next year -either just through changes to the game plan and/or roster/positional changes.
 
The other misleading aspect about the data is that it’s collated based on the ‘season’ and we all know that finals matches are a different ball game altogether. I.e. what works at getting you a win in the regular season does not necessarily correlate to finals wins.

I don’t need anyone to put anything into a computer to know that the risky footy we played in the finals backfired against Souths and Melbourne. We need to play a more % based style of footy to trouble the good sides.
 
The other misleading aspect about the data is that it’s collated based on the ‘season’ and we all know that finals matches are a different ball game altogether. I.e. what works at getting you a win in the regular season does not necessarily correlate to finals wins.

I don’t need anyone to put anything into a computer to know that the risky footy we played in the finals backfired against Souths and Melbourne. We need to play a more % based style of footy to trouble the good sides.
Clearly the game plan was very flawed.

Might have worked , but the odds on it working against the top 3 teams are very low.

If that was Des’s plan then we can hardly blame the players,
 
Lies, lies and damned statistics.

Statistics are not absolute proof of what will happen but rather, only indicative of what may happen.

Medical Research Scientists use it all the time for medical evidence and it is how they identified smoking is bad for you. The correlation of increased lung cancer among smokers versus those who did not smoke … simple.

However, there is also a significant percentage of smokers who have never contracted lung cancer and have just died of natural causes albeit that in the last 20 years heart disease has also been associated with it along with other diseases. So, there is significant medical research proof that if you smoke, you significantly increase the risk of death from lung cancer, heart disease and other ailments.

Statistical correlation and regression do indicate potential cause and effect and particularly in business, it works however, the sample has to be statistically reliable (over n= 858 for business and n=1000 for community) and even then not for all, as there are so many other unexpected variables that can impact on the statistical model. COVID is one example.

Des was, I believe, was the first coach to use a statistician to calculate a ‘type of Super Coach score’ for each player in each game to analyse performance rather than just base it on what he saw and personally I think it’s smart and it’s worked for him. i.e who on this site would have imagined Des would have selected Parker and Harper over Suli and Walker at the beginning of the year (barring suspensions/injuries)? Statistics would have helped him make the decision.

Suli was always a risk, as while he is potentially dynamic in attack (mash them I think one poster said) his errors and defence are his serious weaknesses, as is his ability to set up and pass. So, not only does Suli provide a potential weakness in his position but also the potential weakness in other positions, as they try to cover for him or receive less opportunities. This Is not an excuse for my model however, I didn’t factor Suli in my statistical model of 18-12 prediction with a possible 6 point variable and nor am I bothered to do so however, I don’t think it would have been 36-16. Possibly another 6 point variable, maybe less.

As for Des apparently pissing off players with Suli leaving, all the players mentioned are managed by Tartak (Crown Sports) and you’d think after how Manly have supported Nase Fainu (also Tartak managed) and how Manly have supported the Titmus family (also Tartak managed), he’d have some respect and not try and angle contract upgrades and cause disruption to our great club.

The more I hear about Tartak and Crown Sports the more I think they are ‘grubs’.
 
Clearly the game plan was very flawed.

Might have worked , but the odds on it working against the top 3 teams are very low.

If that was Des’s plan then we can hardly blame the players,
I remember Des interviewed Willie Mason for a gig at the Dogs and told him that Aiden Tolman was the best prop in the game (and he had a book of statistics to prove it). Mason said he laughed and knew then, that he wasn’t gonna sign with the Dogs. The fat lump of **** ended up finishing his career at Manly but he probably a point lol!
 
Yep, you are right. Des rated Tolman the best prop in the comp and I was afraid he’d sign him when he was off contract.

However, if you see Tolman’s stats for metres, post metres, minutes played, tackles made, tackles missed, penalties conceded and how he slowed down the play the ball with aggressive tackles added to his low mistakes, you’ll also see why Des loved Williamson.

Tolman was never a star, but he was a workhorse that you can rely upon. You always need at least one workhorse in every team. Unsung but a huge input.
 
Interesting read, and errors may not be as a large an issue as first thought, our issue isn't so much the errors as closing down the play when we make an error. We seem to all nod off and let the other team just run half the field from our dropped ball
 

Latest posts

Team P W L PD Pts
9 8 1 116 18
9 7 2 72 16
9 7 2 49 16
9 6 3 57 14
10 6 4 115 12
10 6 4 58 12
9 5 4 -14 12
10 5 4 31 11
9 4 5 19 10
10 5 5 -13 10
10 5 5 -56 10
10 4 6 -18 8
9 3 6 -71 8
10 3 6 -9 7
9 2 7 -69 6
9 2 7 -87 6
9 1 8 -180 4
Back
Top Bottom