Birdseye
Well-known member
I wrote this explanation of the system in response to the mis perceptions that there are of NHC. I checked the detail at the time with both the RYA and RORC so it should be correct if simplified. The keen IRC man who is contemptuous of NHC might like to read for a more nuanced reaction.
What's wrong with NYC?
I think it's fair to say that NYC hasn't been received with universal joy . Indeed, not too long after it was introduced when chatting with a table of at least three quarters of the regular racers in a series, every one of those present reckoned that the new handicaps put their boat at a real disadvantage to everyone else ! Think about it ?
This short note explains the way NHC works, some of its faults as well as its advantages and will hopefully counter some of the common misconceptions. So how does it work?
Base NHC handicaps
To devise the individual base handicaps , the RYA turned for help to the RORC who calculate the IRC handicaps . Together they devised a simplified version of the IRC calculation algorithm using basic boat dimensions to predict boat performance. It needed to be a simplified system, because unlike the IRC, the RYA had to cater for a huge range of sometimes quite old boats for which only limited and sometimes unreliable dimension data was available.
When statisticians want to compare two sets of data they calculate something called a correlation coefficient. This is a number varying from +1 through zero to -1 . A correlation of +1 is two sets of numbers marching in lock step whilst zero is two sets of numbers that are totally unrelated. Doing the calculation for IRC and NHC gives a very high 0.9 which proves a hugely close tie between the two handicaps. Interestingly, the correlation between NHC and Byron is much worse at 0.125
Handicap adjustment
NHC is a performance based system just as PY was and for example a golf or horse racing handicap is. So what the RYA then needed was a standard method of adjusting the individual handicaps following each race in a series . The maths of the system chosen isn't that complicated but I doubt that many people want to know the equations involved. So here instead is a simple summary of the way the system works and all you need to know is the significance of a what is called a “standard deviation”..Again its a calculation and its significance is that in a series of random events, like good or bad performance in a race, there is only a 1 in three chance of an event being more than one standard deviation from the mean.
So the computer program called HAL works as follows:
1/ After each race, the finishing order is worked out in the usual way by multiplying a boats elapsed time by its handicap at the at the start to give a corrected time, and then ranking the fleet on the basis of these corrected times.
2/ HAL then works out a handicap number for each finisher which would have meant every boat finishing in a dead heat.
3/ The next step is to calculate your handicap for the next race which HAL does by blending your starting handicap with your handicap under 2 above. But in doing this it only takes into account 30% of the difference. As an illustration, if your start handicap was 0.9 and your handicap from 2/ above was 0.93, HAL will give you a handicap of 0.91 for the next race.
4// But what about those boats who for unusual reasons do particularly badly or wildly well in just that one race? The extreme performers who are more than one standard deviation from the mean. It could be, for example, that after half the fleet had finished the wind died to near nothing. So to prevent excessive large adjustments, HAL trets these boats as if they had finished at the time corresponding to one standard deviation.
5/ The final step that HAL takes is one to prevent the drift up or down of all the handicaps in a fleet. It does this in effect by adjusting all the new post race handicaps so that on average they are close to the base NHC handicap number. This doesnt alter the relativities between boats – they all move up or down together including those who didnt even turn out .
The misconceptions
So let's deal with a few, together with some of the strengths and weaknesses of NHC:
1/ The race officer can fiddle the handicaps and look after his pals. Leaving aside that we don't have race officers like that, no he can't. The post race handicap adjustment is computer generated.
2/ Bad sailors can win in NHC. No they can't unless they start as bad sailors and significantly improve. NHC rewards those who do just a bit better every race. Even then it takes a long time for the handicap starting at the base number to fall with bad performance until the boat starts finishing in the top half and the handicap starts climbing again. Remember the 30%
3/ You can game NHC by sailing badly until your handicap is low and then sailing well. To a very limited extent, you can cheat in this way, just as you can with self measurement in IRC. Or indeed by missing marks, or hiring hot shot crew.
4/ NHC punishes you for sailing well. Yes it takes 30% of your out performance away from you each race, so you have to do a bit better each time. But it does avoid the situation where the same boat is winning all the time and the losers simply can never compete.
5/ NHC base numbers are over simple and it takes time for a handicap to reflect reality. True. For some reason the RYA dropped the old PY adjustment factors for things like twin keels. small sails, 3 blade props.
There are Pro and Con for the two basic handicap types, and the proponents of each need to recognise that they are designed for different types of racer. In the absence of a pure one make fleet, the keen racers will nearly always go for a design based handicap like IRC recognising that to a significant degree he will have to email and man his boat to the appropriate standard if he wishes to win. A family sailor in a non racing type of boat and simply wanting some Sunday fun will mostly code a performance based system that will still give him a real chance even if he is sailing a 30 year old bilge Keeler with its original sails and all his cruising gear on board. Horses for courses.
What is true of either handicap system is that they only really work well with similar boats. No handicap that will cover an Impala and a J109 in 10 kn wind and a neap tide will work in 25kn breeze and a spring tide.
What's wrong with NYC?
I think it's fair to say that NYC hasn't been received with universal joy . Indeed, not too long after it was introduced when chatting with a table of at least three quarters of the regular racers in a series, every one of those present reckoned that the new handicaps put their boat at a real disadvantage to everyone else ! Think about it ?
This short note explains the way NHC works, some of its faults as well as its advantages and will hopefully counter some of the common misconceptions. So how does it work?
Base NHC handicaps
To devise the individual base handicaps , the RYA turned for help to the RORC who calculate the IRC handicaps . Together they devised a simplified version of the IRC calculation algorithm using basic boat dimensions to predict boat performance. It needed to be a simplified system, because unlike the IRC, the RYA had to cater for a huge range of sometimes quite old boats for which only limited and sometimes unreliable dimension data was available.
When statisticians want to compare two sets of data they calculate something called a correlation coefficient. This is a number varying from +1 through zero to -1 . A correlation of +1 is two sets of numbers marching in lock step whilst zero is two sets of numbers that are totally unrelated. Doing the calculation for IRC and NHC gives a very high 0.9 which proves a hugely close tie between the two handicaps. Interestingly, the correlation between NHC and Byron is much worse at 0.125
Handicap adjustment
NHC is a performance based system just as PY was and for example a golf or horse racing handicap is. So what the RYA then needed was a standard method of adjusting the individual handicaps following each race in a series . The maths of the system chosen isn't that complicated but I doubt that many people want to know the equations involved. So here instead is a simple summary of the way the system works and all you need to know is the significance of a what is called a “standard deviation”..Again its a calculation and its significance is that in a series of random events, like good or bad performance in a race, there is only a 1 in three chance of an event being more than one standard deviation from the mean.
So the computer program called HAL works as follows:
1/ After each race, the finishing order is worked out in the usual way by multiplying a boats elapsed time by its handicap at the at the start to give a corrected time, and then ranking the fleet on the basis of these corrected times.
2/ HAL then works out a handicap number for each finisher which would have meant every boat finishing in a dead heat.
3/ The next step is to calculate your handicap for the next race which HAL does by blending your starting handicap with your handicap under 2 above. But in doing this it only takes into account 30% of the difference. As an illustration, if your start handicap was 0.9 and your handicap from 2/ above was 0.93, HAL will give you a handicap of 0.91 for the next race.
4// But what about those boats who for unusual reasons do particularly badly or wildly well in just that one race? The extreme performers who are more than one standard deviation from the mean. It could be, for example, that after half the fleet had finished the wind died to near nothing. So to prevent excessive large adjustments, HAL trets these boats as if they had finished at the time corresponding to one standard deviation.
5/ The final step that HAL takes is one to prevent the drift up or down of all the handicaps in a fleet. It does this in effect by adjusting all the new post race handicaps so that on average they are close to the base NHC handicap number. This doesnt alter the relativities between boats – they all move up or down together including those who didnt even turn out .
The misconceptions
So let's deal with a few, together with some of the strengths and weaknesses of NHC:
1/ The race officer can fiddle the handicaps and look after his pals. Leaving aside that we don't have race officers like that, no he can't. The post race handicap adjustment is computer generated.
2/ Bad sailors can win in NHC. No they can't unless they start as bad sailors and significantly improve. NHC rewards those who do just a bit better every race. Even then it takes a long time for the handicap starting at the base number to fall with bad performance until the boat starts finishing in the top half and the handicap starts climbing again. Remember the 30%
3/ You can game NHC by sailing badly until your handicap is low and then sailing well. To a very limited extent, you can cheat in this way, just as you can with self measurement in IRC. Or indeed by missing marks, or hiring hot shot crew.
4/ NHC punishes you for sailing well. Yes it takes 30% of your out performance away from you each race, so you have to do a bit better each time. But it does avoid the situation where the same boat is winning all the time and the losers simply can never compete.
5/ NHC base numbers are over simple and it takes time for a handicap to reflect reality. True. For some reason the RYA dropped the old PY adjustment factors for things like twin keels. small sails, 3 blade props.
There are Pro and Con for the two basic handicap types, and the proponents of each need to recognise that they are designed for different types of racer. In the absence of a pure one make fleet, the keen racers will nearly always go for a design based handicap like IRC recognising that to a significant degree he will have to email and man his boat to the appropriate standard if he wishes to win. A family sailor in a non racing type of boat and simply wanting some Sunday fun will mostly code a performance based system that will still give him a real chance even if he is sailing a 30 year old bilge Keeler with its original sails and all his cruising gear on board. Horses for courses.
What is true of either handicap system is that they only really work well with similar boats. No handicap that will cover an Impala and a J109 in 10 kn wind and a neap tide will work in 25kn breeze and a spring tide.