Posts from June 2021 (26)

June 15, 2021

Rugby Premiership Predictions for the Semi-finals

Team Ratings for the Semi-finals

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Exeter Chiefs 9.63 7.35 2.30
Sale Sharks 5.91 4.96 0.90
Bristol 5.49 1.28 4.20
Harlequins 2.66 -1.08 3.70
Northampton Saints -0.75 -2.48 1.70
Wasps -1.95 5.66 -7.60
Gloucester -2.57 -1.02 -1.50
Leicester Tigers -2.70 -6.14 3.40
Bath -2.81 2.14 -4.90
London Irish -7.17 -8.05 0.90
Newcastle Falcons -9.04 -10.00 1.00
Worcester Warriors -9.80 -5.71 -4.10

 

Performance So Far

So far there have been 122 matches played, 83 of which were correctly predicted, a success rate of 68%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Bath vs. Northampton Saints Jun 13 30 – 24 1.90 TRUE
2 Exeter Chiefs vs. Sale Sharks Jun 13 20 – 19 9.20 TRUE
3 Harlequins vs. Newcastle Falcons Jun 13 54 – 26 14.80 TRUE
4 Wasps vs. Leicester Tigers Jun 13 31 – 38 6.70 FALSE

 

Predictions for the Semi-finals

Here are the predictions for the Semi-finals. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Bristol vs. Harlequins Jun 20 Bristol 7.30
2 Exeter Chiefs vs. Sale Sharks Jun 20 Exeter Chiefs 8.20

 

Pro 14 Rainbow Cup Predictions for the Final

Team Ratings for the Final

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Leinster 14.79 15.74 -1.00
Munster 10.69 9.13 1.60
Ulster 7.41 8.77 -1.40
Bulls 4.63 1.70 2.90
Glasgow Warriors 3.69 2.64 1.00
Edinburgh 2.90 3.48 -0.60
Connacht 1.72 2.50 -0.80
Ospreys 0.94 -0.65 1.60
Stormers 0.00 0.80 -0.80
Sharks -0.07 1.16 -1.20
Cardiff Blues -0.11 0.73 -0.80
Scarlets -0.77 0.41 -1.20
Lions -3.91 -3.00 -0.90
Treviso -5.48 -6.46 1.00
Dragons -6.92 -6.39 -0.50
Zebre -13.47 -14.52 1.10

 

Performance So Far

So far there have been 39 matches played, 19 of which were correctly predicted, a success rate of 48.7%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Zebre vs. Munster Jun 12 11 – 54 -15.80 TRUE
2 Leinster vs. Dragons Jun 12 38 – 7 27.60 TRUE
3 Sharks vs. Bulls Jun 13 22 – 34 1.40 FALSE
4 Scarlets vs. Edinburgh Jun 13 28 – 28 3.50 FALSE

 

Predictions for the Final

Here are the predictions for the Final. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Treviso vs. Bulls Jun 20 Bulls -3.60

 

NRL Predictions for Round 15

Team Ratings for Round 15

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Storm 18.96 14.53 4.40
Panthers 13.56 8.88 4.70
Eels 9.02 1.68 7.30
Roosters 8.27 10.25 -2.00
Rabbitohs 4.69 7.73 -3.00
Sea Eagles 3.89 -4.77 8.70
Sharks -1.13 -0.76 -0.40
Raiders -2.35 6.98 -9.30
Wests Tigers -2.67 -3.07 0.40
Warriors -3.14 -1.84 -1.30
Dragons -4.91 -4.95 0.00
Titans -5.87 -7.22 1.40
Knights -8.57 -2.61 -6.00
Cowboys -9.59 -8.05 -1.50
Bulldogs -9.76 -7.62 -2.10
Broncos -12.41 -11.16 -1.30

 

Performance So Far

So far there have been 108 matches played, 74 of which were correctly predicted, a success rate of 68.5%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Sea Eagles vs. Cowboys Jun 11 50 – 18 13.20 TRUE
2 Sharks vs. Panthers Jun 11 19 – 18 -14.40 FALSE
3 Titans vs. Roosters Jun 12 34 – 35 -13.30 TRUE
4 Rabbitohs vs. Knights Jun 12 24 – 10 16.80 TRUE
5 Raiders vs. Broncos Jun 12 38 – 16 11.10 TRUE
6 Warriors vs. Storm Jun 13 16 – 42 -21.20 TRUE
7 Eels vs. Wests Tigers Jun 13 40 – 12 11.90 TRUE
8 Bulldogs vs. Dragons Jun 14 28 – 6 -6.70 FALSE

 

Predictions for Round 15

Here are the predictions for Round 15. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Broncos vs. Rabbitohs Jun 17 Rabbitohs -14.10
2 Cowboys vs. Sharks Jun 18 Sharks -5.50
3 Panthers vs. Roosters Jun 18 Panthers 8.30
4 Knights vs. Warriors Jun 19 Warriors -2.40
5 Dragons vs. Raiders Jun 19 Dragons 0.40
6 Storm vs. Wests Tigers Jun 19 Storm 24.60
7 Eels vs. Bulldogs Jun 20 Eels 21.80
8 Titans vs. Sea Eagles Jun 20 Sea Eagles -6.80

 

June 14, 2021

Controlling emissions

There are two basic ways a government can put downwards pressure on carbon emissions and let consumers find the best ways to adapt. One is to charge a fixed tax per kg of CO2 equivalent, the other is to fix a total cap on emissions and let consumers bid for the rights to use it.  There’s a sense in which these are equivalent: if the price in a cap-and-trade system ends up being $X per kg, the emissions will be the same as if the government charged $X per kg and didn’t have a cap. A sufficiently flexible and adaptive version of either one could match the other.  In reality they aren’t quite the same because governments want a simple and relatively predictable price or cap.

We’ve got a cap (more or less). One of the non-intuitive aspects of having a cap rather than a fixed price is that parallel efforts to reduce carbon emission don’t work the way you’d expect them to. If I replace my gas stove with an electric one, my kitchen will emit less carbon (modulo the impacts of making the new equipment).  If everyone did it, everyone’s kitchen would emit less carbon (again, ignoring the impacts of making the new equipment).  What would happen to NZ’s total carbon emissions? Nothing. We have a cap.  Less of the cap would go on carbon coupons for burning natural gas; more of it would be available for cars or trucks or coal-fired power stations.  The impact of our kitchen-renovation decisions would be cheaper emissions rights for other polluters, not lower emissions.

In principle, I could keep buying emissions rights for the natural gas I wasn’t using. That would turn my lower emissions into reductions for NZ as a whole. Or, the government could monitor the sales of induction cooktops and withdraw emissions rights to compensate (or, more realistically, track kitchen conversions through some sort of subsidy).  But if nothing happens to the total ETS carbon budget, nothing will happen to total emissions. A big enough change in demand could change emissions — if cars were suddenly banned, the government might not be able to sell all its ETS coupons — but a modest change won’t.

When the government says that new subsidies for low-emissions cars will reduce carbon emissions by some large number, there’s a gap in the explanation.  Having more low-emissions cars will lower carbon emissions by cars, but unless the government withdraws the corresponding emissions rights from the carbon budget, it won’t reduce carbon emissions in total. The reduction will go to lowering carbon costs for other polluters.

This, in itself, doesn’t mean the policy is bad; it just means the policy needs to be evaluated in some other way.  Maybe subsidising electric cars will lower the cost of future emissions reductions. Maybe it will improve the political feasibility of reducing the total emissions budget. Maybe there’s some other big benefit that I haven’t seen. But it is a problem that the policy is being sold on emissions reductions and that there doesn’t seem to be media or political reaction asking exactly where these reductions are coming from given the ETS cap.

New Alzheimer’s drug

The US Food and Drug Administration has approved a new treatment, aducanumab for Alzheimer’s Disease. The Chief Executive of Alzheimers NZ was quoted by  TVNZ

“What I’d say is, cautious optimism,” Alzheimer’s New Zealand chief executive Catherine Hall told 1 NEWS when asked about her reaction to the drug’s approval overseas.

She said people living with dementia often ride a rollercoaster of emotion when it comes to new drugs being announced.

“They get told there’s a brand new cure and then very quickly afterwards their hopes are dashed. It’s really important to recognise this is early days and there’s still a lot of data to be collected.”

It’s a very sensible attitude, in the abstract: if the drug turns out to be effective it could be valuable, but it’s too early to know if that will be the case. What’s surprising is that this is the situation we’re in after the drug has been approved, and when its manufacturer is planning to charge US$56000/year for it.

The drug (or, technically, the ‘biologic’ since it’s an antibody) has been through a lot of ups and downs in its clinical trial history.  There were two main trials that were supposed to show it was effective. They failed. A re-analysis of one of them suggested that it might actually work, at least for some patients. Normally, this would be the cue to do a confirmatory trial to see if it does actually help an identifiable group of people. And the FDA did mandate this trial — but they will let the manufacturer sell and promote the medication for nine years while the trial goes on.  Given that the the market for aducanumab is conservatively estimated at tens of millions of dollars per day, and there’s only a possible downside to getting trial results, the trial is unlikely to end a day sooner than it has to; it’s not unheard of for these post-approval trials to just never recruit enough participants and drag on longer than ‘allowed’.

The FDA takes external expert advice on drug approvals. In this case, there were 11 people on the panel. Exactly none of them thought there was good enough evidence for approval; one was uncertain, ten were against. Three of the panel members have since resigned. It’s not unprecedented for the FDA to disagree with the panel when the panel vote is split, but it’s pretty bloody unusual for them to disagree with a unanimous panel.  It’s notable that the FDA approval does not say they think there’s evidence drug improves memory or cognition or ability to live independently or anything like that — the FDA thinks it reduces the amount of amyloid plaque in patients’ brains and hopes this will translate to (currently unobserved) improvements in how they live.

TVNZ goes on to say

It’s only suitable for those with mild Alzheimer’s when the damage to brain function is still limited.

This is what the trials would say, under the optimistic interpretation that they say it’s suitable for anyone. The FDA, however, says it is an amyloid beta-directed antibody indicated for the treatment of Alzheimer’s disease. No restriction. Since most people who’d want the drug are over 65, it will predominantly be funded by Medicare, the US government health program. Medicare tends to pay for anything the FDA approves, but they might not have the option this time. Presumably in NZ we won’t get Pharmac subsidy for aducanumab at least until it’s shown to work; an incautiously  optimistic date would be 2030.

Something like the aducanumab approval is what many of us were afraid of for Covid vaccines — an product with weak evidence of modest effectiveness being given the green light because demand was high: “something must be done; this is something; therefore we must do it”.  We were lucky with Covid that the temptation didn’t arise for the FDA: the vaccines it has been asked to approve are more effective than anyone could have hoped.  Aducanumab won’t be that sort of disaster, but it will put a lot of pressure on a US health cover system for older people where the government is expected to pay but not to count the cost, let alone balance it against effectiveness.

June 8, 2021

Top 14 Predictions for the Semi-finals

Team Ratings for the Semi-finals

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
La Rochelle 7.30 2.32 5.00
Racing-Metro 92 7.05 6.21 0.80
Stade Toulousain 7.03 4.80 2.20
Clermont Auvergne 5.54 3.22 2.30
Bordeaux-Begles 4.65 2.83 1.80
Lyon Rugby 4.55 5.61 -1.10
RC Toulonnais 1.91 3.56 -1.70
Stade Francais Paris 1.59 -3.22 4.80
Castres Olympique 1.29 -0.47 1.80
Montpellier 0.52 2.30 -1.80
Section Paloise -2.55 -4.48 1.90
Brive -3.29 -3.26 -0.00
Aviron Bayonnais -5.82 -4.13 -1.70
SU Agen -19.21 -4.72 -14.50

 

Performance So Far

So far there have been 182 matches played, 127 of which were correctly predicted, a success rate of 69.8%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Aviron Bayonnais vs. Stade Francais Paris Jun 06 9 – 12 -1.70 TRUE
2 Bordeaux-Begles vs. Stade Toulousain Jun 06 10 – 21 4.20 FALSE
3 Castres Olympique vs. RC Toulonnais Jun 06 46 – 24 3.60 TRUE
4 Clermont Auvergne vs. La Rochelle Jun 06 25 – 20 3.50 TRUE
5 Lyon Rugby vs. SU Agen Jun 06 52 – 7 28.10 TRUE
6 Racing-Metro 92 vs. Brive Jun 06 55 – 12 14.10 TRUE
7 Section Paloise vs. Montpellier Jun 06 41 – 25 1.40 TRUE

 

Predictions for the Semi-finals

Here are the predictions for the Semi-finals. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Racing-Metro 92 vs. Stade Francais Paris Jun 12 Racing-Metro 92 11.00
2 Bordeaux-Begles vs. Clermont Auvergne Jun 12 Bordeaux-Begles 4.60

 

Super Rugby Predictions for Week 17

Team Ratings for Week 17

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Crusaders 14.46 14.49 -0.00
Blues 9.87 7.80 2.10
Hurricanes 7.46 7.13 0.30
Highlanders 4.61 2.70 1.90
Chiefs 4.53 4.38 0.20
Brumbies 2.83 1.47 1.40
Reds 1.09 1.59 -0.50
Rebels -6.06 -3.51 -2.50
Waratahs -10.18 -5.02 -5.20
Western Force -10.64 -13.05 2.40

 

Performance So Far

So far there have been 63 matches played, 50 of which were correctly predicted, a success rate of 79.4%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Crusaders vs. Western Force Jun 04 29 – 21 33.20 TRUE
2 Reds vs. Blues Jun 04 24 – 31 -1.60 TRUE
3 Highlanders vs. Waratahs Jun 05 59 – 23 20.20 TRUE
4 Brumbies vs. Hurricanes Jun 05 12 – 10 1.80 TRUE
5 Chiefs vs. Rebels Jun 06 36 – 26 10.70 TRUE

 

Predictions for Week 17

Here are the predictions for Week 17. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Hurricanes vs. Reds Jun 11 Hurricanes 12.90
2 Brumbies vs. Highlanders Jun 11 Brumbies 4.70
3 Rebels vs. Crusaders Jun 12 Crusaders -20.50
4 Blues vs. Western Force Jun 12 Blues 27.00
5 Waratahs vs. Chiefs Jun 12 Chiefs -8.20

 

Rugby Premiership Predictions for Round 22

Team Ratings for Round 22

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Exeter Chiefs 10.11 7.35 2.80
Bristol 5.49 1.28 4.20
Sale Sharks 5.44 4.96 0.50
Harlequins 1.95 -1.08 3.00
Northampton Saints -0.48 -2.48 2.00
Wasps -1.21 5.66 -6.90
Gloucester -2.57 -1.02 -1.50
Bath -3.08 2.14 -5.20
Leicester Tigers -3.44 -6.14 2.70
London Irish -7.17 -8.05 0.90
Newcastle Falcons -8.32 -10.00 1.70
Worcester Warriors -9.80 -5.71 -4.10

 

Performance So Far

So far there have been 118 matches played, 80 of which were correctly predicted, a success rate of 67.8%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Sale Sharks vs. Harlequins Jun 05 45 – 12 5.30 TRUE
2 Leicester Tigers vs. Bristol Jun 06 23 – 26 -4.70 TRUE
3 London Irish vs. Wasps Jun 06 36 – 39 -1.20 TRUE
4 Newcastle Falcons vs. Worcester Warriors Jun 06 24 – 14 5.40 TRUE
5 Northampton Saints vs. Exeter Chiefs Jun 07 26 – 29 -6.70 TRUE

 

Predictions for Round 22

Here are the predictions for Round 22. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Bath vs. Northampton Saints Jun 13 Bath 1.90
2 Bristol vs. London Irish Jun 13 Bristol 17.20
3 Exeter Chiefs vs. Sale Sharks Jun 13 Exeter Chiefs 9.20
4 Harlequins vs. Newcastle Falcons Jun 13 Harlequins 14.80
5 Wasps vs. Leicester Tigers Jun 13 Wasps 6.70
6 Worcester Warriors vs. Gloucester Jun 13 Gloucester -2.70

 

Pro 14 Rainbow Cup Predictions for Week 6

 

 

Team Ratings for Week 6

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Leinster 14.49 15.74 -1.30
Munster 9.78 9.13 0.70
Ulster 7.41 8.77 -1.40
Bulls 4.08 1.70 2.40
Glasgow Warriors 3.69 2.64 1.00
Edinburgh 2.58 3.48 -0.90
Connacht 1.72 2.50 -0.80
Ospreys 0.94 -0.65 1.60
Sharks 0.48 1.16 -0.70
Stormers 0.00 0.80 -0.80
Cardiff Blues -0.11 0.73 -0.80
Scarlets -0.46 0.41 -0.90
Lions -3.91 -3.00 -0.90
Treviso -5.48 -6.46 1.00
Dragons -6.62 -6.39 -0.20
Zebre -12.56 -14.52 2.00

 

Performance So Far

So far there have been 35 matches played, 17 of which were correctly predicted, a success rate of 48.6%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Bulls vs. Stormers Jun 05 37 – 12 7.80 TRUE
2 Connacht vs. Ospreys Jun 05 21 – 33 8.80 FALSE
3 Glasgow Warriors vs. Leinster Jun 05 31 – 34 -4.60 TRUE
4 Cardiff Blues vs. Zebre Jun 06 31 – 34 20.60 FALSE
5 Edinburgh vs. Ulster Jun 06 37 – 12 -0.00 FALSE
6 Lions vs. Sharks Jun 06 21 – 33 1.70 FALSE

 

Predictions for Week 6

Here are the predictions for Week 6. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Zebre vs. Munster Jun 12 Munster -15.80
2 Leinster vs. Dragons Jun 12 Leinster 27.60
3 Stormers vs. Lions Jun 13 Stormers 8.90
4 Sharks vs. Bulls Jun 13 Sharks 1.40
5 Ospreys vs. Treviso Jun 13 Ospreys 12.90
6 Scarlets vs. Edinburgh Jun 13 Scarlets 3.50

 

NRL Predictions for Round 14

Team Ratings for Round 14

The basic method is described on my Department home page.
Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Current Rating Rating at Season Start Difference
Storm 18.52 14.53 4.00
Panthers 14.91 8.88 6.00
Roosters 9.35 10.25 -0.90
Eels 7.62 1.68 5.90
Rabbitohs 4.98 7.73 -2.80
Sea Eagles 2.27 -4.77 7.00
Wests Tigers -1.27 -3.07 1.80
Sharks -2.48 -0.76 -1.70
Dragons -2.49 -4.95 2.50
Warriors -2.70 -1.84 -0.90
Raiders -3.31 6.98 -10.30
Titans -6.96 -7.22 0.30
Cowboys -7.97 -8.05 0.10
Knights -8.85 -2.61 -6.20
Broncos -11.44 -11.16 -0.30
Bulldogs -12.18 -7.62 -4.60

 

Performance So Far

So far there have been 100 matches played, 68 of which were correctly predicted, a success rate of 68%.
Here are the predictions for last week’s games.

Game Date Score Prediction Correct
1 Dragons vs. Broncos Jun 03 52 – 24 8.60 TRUE
2 Wests Tigers vs. Panthers Jun 04 26 – 6 -19.80 FALSE
3 Storm vs. Titans Jun 05 20 – 14 33.10 TRUE
4 Knights vs. Eels Jun 06 4 – 40 -8.90 TRUE

 

Predictions for Round 14

Here are the predictions for Round 14. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Game Date Winner Prediction
1 Sea Eagles vs. Cowboys Jun 11 Sea Eagles 13.20
2 Sharks vs. Panthers Jun 11 Panthers -14.40
3 Titans vs. Roosters Jun 12 Roosters -13.30
4 Rabbitohs vs. Knights Jun 12 Rabbitohs 16.80
5 Raiders vs. Broncos Jun 12 Raiders 11.10
6 Warriors vs. Storm Jun 13 Storm -21.20
7 Eels vs. Wests Tigers Jun 13 Eels 11.90
8 Bulldogs vs. Dragons Jun 14 Dragons -6.70