NHC Ratings used

Progressive Handicap
1

Having read so much about the NHC scoring system, I created a copy of our last summer series, created alias’ for each competitor and assigned to use the NHC system. It seems to work except for the PHRF ratings. We are in So Cal and it is a small area using only PHRF.
I failed to understand the rating the system (not PHRF) and got some wonky results. I looked up the RYA base sheet and found a few or the boats listed (J boats, Bene 25, C&C110, Moore 24), but most of the fleet is no listed.
What I really need know is what system is used? or the equation used for NHC ratings.
How about a conversion calculator so I can take a PHRF rating and get a usable number for the RYA NHC scoring.
NHC rating looks a lot like an ORR rating. But ORR has many different ratings based on the course and wind strength.

Thanks.

2

NHC uses time-correction factor (TCF) calculations; these are sometimes
referred to as Time-on-Time or ToT calculations. The correction formula
is Corrected Time = Elapsed Time x TCF. The RYA issues ToT (aka TCF)
ratings and the NHC system is set to use ToT/TCF ratings.

SoCal PHRF issues ratings as Time-on-Distance (ToD) ratings. So, you’d
have to convert the SoCal PHRF ToD ratings into ToT ratings to use with
the NHC system.

I believe 99.9% of rating systems fall into either ToD or ToT.

There is a standard formula for the conversion of ToD to ToT:

ToT TCF = STD/(STD+PHRF) [you often see A and B but I’ll explain that below]

STD is the number of seconds per mile that a boat with a PHRF rating of
0 will take to sail 1 mile in average wind conditions for the area where
the ratings were issued. The idea is that the TCF for a 0-rated PHRF
boat will be 1.000 since the numerator and denominator will be the same
(STD = STD+0). Faster boats will have a larger TCF and slower boats will
have a lower TCF. A lower TCF results in a smaller corrected time.

Except in rare circumstances I think you always want a 0-rated boat to
have a TCF of 1.000 and to get that you want the top factor and the
bottom factor to be the same. You often see the conversion formula as
TCF = A/(B+PHRF) implying that the factors should be different; however
all that is doing is changing which boats will have a TCF of 1.000. The
A factor has no impact on the order of corrected times for boats
(assuming it isn’t 0, making division impossible); all it does is reduce
the gap between the corrected times of boats by making the time for the
whole race shorter (reducing the gaps) or longer (increasing the gaps).
For consistency between ToD and ToT I believe a 0-rated boat should
always have its corrected time equal to its elapsed time regardless of
whether ToD or ToT ratings are used. That should be easier for
competitors to understand.

I am in an area with light to moderate winds (Marina del Rey in Los
Angeles, CA if you want to know) and for our area the average time of a
0-rated boat is 630-650 seconds per mile. If you were in a windier area,
like Santa Barbara, CA you would probably want to use a number around
600. In an area where the average wind is 20+ (like San Francisco, CA) I
would consider using something around 580. Those are numbers for ratings
for windward/leeward courses. If the course would be mostly reaching
then the numbers are going to be significantly smaller (with speed
expressed as seconds/mile, a smaller number is faster).

You can confirm what kind of number to use for STD by taking the elapsed
times for the middle boat for 10 typical races in your area, dividing by
the course distance and subtracting the rating for the boat. EXAMPLE:
the middle boat, rating 60, sails a 5-mile race in 3500 seconds. 3500/5
= 700 seconds/mile for a 60-rated boat = 640 seconds/mile for a 0-rated
boat. Your STD would be 640.

Art

3

Thanks Art, I’m up in Area A (Ventura, Channel Islands). I’m sure we have met at some point. I was on the PHRF board for several years as an area rep. My boat is a Zap 26 56267, Gumbercules.
Our area during the summer Wet Wednesday can be pretty windy but in last month of the series it can get pretty light and shifty. I’ll probably just use the suggested numbers as they probably cover the range.
I have been doing the scoring here for too many years and I think the RYA NHC progressive scoring will make it much easier on the volunteers that follow.

I understand the ToD and ToT along with the TCF calc. I just did not realize that the RYA NHC was using the same TCF that ToT was using. I should have titled this thread Alphabet Soup.
All it good. Thanks

4

OK, so I used a J35-72, My Zap26-135, J24-168 and averaged an STD of 612 for A and 550 (average conditions) for B. I used some older races to calculate the 612, where we had bigger fleets
If I did it right the TCF = 612 / (550+PHRF)
after Calculating the small fleet and entering the TCH for each boat the scores look much better.
Attached is the .blw file.

The file has the original PHRF ToD scores, and the Alias fleets are scored using the NHC. One thing I looked at was Rival, the J35 in the first 6 races. It looks like his rating is getting hit but then it gets bumped up. Is this because there is only 3 boats in the first races?

Someone tell me I did it right and can present it to next year’s Bridge for review.

z RYA NHC 2022 WW.blw (227.5 KB)

5

No, an STD of 612 is the B factor on the bottom.

The A factor on the top doesn’t impact the order of boats, it just
determines which boat will have a TCF/ToT of 1.000. For an A factor you
would use the B factor plus the PHRF rating for the boat you want to
have a TCF/ToT of 1.000. If you want a boat with a PHRF rating of 120 to
have a TCF/ToT of 1.000 then you’d use 612+120 = 732 as your A factor.

In my experience, competitors used to standard PHRF ToD ratings have a
very difficult time understanding races results scored with TCF/ToT
instead (your own experience might be a good example of that).
Therefore, I would HIGHLY recommend that you use an A factor of B+0,
which means that for a 0-rated PHRF boat scored with TCF/ToT ratings the
corrected time will be the same as the elapsed time, just like with
standard PHRF ToD ratings. That should be easier for competitors to
understand - easier to understand but not to imply that they actually
will understand!

A B factor of 550 would imply that boats sail around the course at a
VERY high average speed. I think that would imply a 100%
reaching/running course - reaching out and reaching back or downwind
only - with no windward legs. That would be way too low a number for a
typical windward/leeward buoy race, in my opinion.

A B factor of 612 sounds about right (maybe slightly high) for the
Ventura/Channel Islands area. I think a B factor of 550 for buoy races
might kind of imply all your boats can plane upwind, certainly not
possible with a displacement boat like a J/35.

To check that the TCF/ToT ratings you are generating make sense then
just rescore the races using those ratings. For races with average wind
strength, the TCF/ToT ratings should generate results very similar to
standard PHRF ToD scoring. If not then your conversion factors are not
right. TCF/ToT scoring should only produce a different order of
corrected times when the wind conditions are a LOT different from
average - much lighter or much, much heavier.

As for the NHC system, I am not a big fan. Some of the calculations
don’t make sense to me. So, I cannot advise you on that. But, I think
you want to make sure that what you are comparing is the non-NHC TCF/ToT
results with the NHC results. You don’t want to compare apples with
oranges by using results generated with standard PHRF ToD ratings.

One thing to consider, it might be possible to customize the NHC
calculations so that you input ToD ratings and it generates revised ToD
ratings. That would be certainly be easier for competitors to
understand. But, it would require some manipulation of an external Excel
spreadsheet.

I’ve never attempted that with Sailwave although I have played around
with various ways to calculate skipper-adjustments (in other words, an
NHC-type system) for local races using Excel. In the end, I decided
skipper handicaps have too many problems except in novice racing.
Locally, the boat the won our Wed. night series each year simply lost
the first 3 races by big margins and then always finished 2nd in the
remaining races. In theory, NHC tries to prevent that by excluding
results that are too good or too bad but I’m sure it is still subject to
significant manipulation if someone wants to take the time to bother.
Plus, while such a system allows the newbie sailors to get the
occasional trophy it rewords poor sailing, which I personally kind of
think might be the wrong incentive.

Art

6

Dave,

Your file seems to be messed up.

  1. The ratings for your standard non-NHC boats are changing for every
    race. That shouldn’t happen. I think it means your non-NHC results are
    meaningless. It seems you have set it up so that each boat’s rating
    changes after each race. That means your “standard” results also use
    skipper handicapping, just a different kind of skipper handicaps from
    the NHC system.

  2. Your starting ratings for the NHC classes are wrong. Example: DJ
    (J/70) starts with a rating of 1.08 (by the way, you should use ratings
    to 3 decimal places to have approx. 1 sec/mi accuracy). The formula you
    said you used was TCF=612/(550+PHRF). But, that means the the rating for
    DJ should be 612/(550+114)=612/664=0.922 not 1.080. The slower boats
    have bigger TCF/ToT ratings when they should be smaller. So, somewhere
    you probably divided the wrong numbers.

I kind of think you don’t have enough boats to do skipper handicapping
using the NHC system. One class averages 4 boats while the other two
classes are averaging 2 boats in a typical race. NHC needs a minimum of
3 boats and excludes boats that are too fast or too slow (are outside
some percentage of a standard deviation). I think that is going to leave
you with too few boats for meaningful adjustments under the NHC system,
though I could be wrong (certainly you could generate adjustments but I
don’t think they’d really be meaningful because of the small sample
size). I think you’d be better with some kind of a simple formula.

Locally, we used to have skipper handicapping (we no longer do). We
measured how much faster or slower (in secs/mi) each boat was than the
boat correcting as the middle boat for that race. We then adjusted the
rating for the next race by a percentage (10-15%, it changed over the
years) of that faster/slower amount, with a maximum change limited to
some number (I think it was 30 sec/mi). I believe NHC compares to the
1st-place boat which makes for a distortion which they try to correct by
“re-indexing” all ratings after each race (their system could be
improved by comparing to the middle boat). If I remember correctly that
is why the ratings for boats that don’t race can change whereas under
the system we used boats that don’t race never change ratings. We found
that was easier for competitors to understand.

Good sailing,
Art

7

I have a question re using NHC. I took Dave’s BLW file and played
around. One thing I did was “try” NHC by changing to the RYA NHC scoring
system. All well and good. Then I changed back to my normal system of
PHRF ToD and scored (not rescored).

BUT, now each competitor had a “race ratings” that was different for
each race - presumably from having used NHC. So, the base ratings I
entered for each competitor apparently weren’t being used; instead there
was now a new “race rating” being used. I assumed that when I switched
away from NHC that any new race ratings assigned by NHC would disappear.
And, I can’t figure out how to get rid of those race ratings except by
going through every boat for every race and using “Edit result” manually
deleting the race rating (so that without a race rating my base ratings
would be used).

Is there any way to remove the race ratings generated by NHC when I have
selected another scoring system? If I had known about the problem I
might have saved a copy just before experimenting with NHC. For me this
is just a test case but this could be a significant problem if it
happened on race day with an actual scoring file.

By the way, I looked at “set competitors” but that doesn’t seem to give
me the option to set the race rating field. I think my problem would be
solved if I could simply set all race ratings as blank. There is a “new
rating” field but setting it to be blank doesn’t seem to do anything
with respect to “race ratings.” I note that in a normal file the “race
rating” fields are blank.

Art

8

the non NHC PHRF boats are how we do it all summer long. 1st place gets a -6, 2nd -3, Last place gets a +6, 2nd from last gets +3. It is fraught with human errors especially when the rum comes out. :crazy_face:
I wanted to see how NHC scored compared to the manual way we have been doing it. It is a pain in the ass to track ratings all summer long, 23 races. No one Volunteers for scoring anymore, so I am trying to simplify the scoring with a progressive system.

I screwed the pooch on the division, sometimes it is good to take a break. I’ll program a spread sheet and make it simple.
I wish I could score it as a group but sometimes they send the smaller boats on a different course. The C fleet is the cruising NonSpin fleet, we ger very few there.

Thanks for catching my simple math mistake

9

I got this off of a PHRF Site
|B Factor|When Used|
|480|Heavy air or all off the wind|
|550|Average conditions|
|600|Very light air or all windward work|

The formula to figure the A factor I found is:
(Elapsed Time (seconds) / Race Distance) - Phrf Rating

The 612 A factor was calculated and averaged from 3 of the best kept and sailed boats in the area. Usually, the ones making the least mistakes too. So it is based on Good Sailors, Good Sails, Good starts, not overstanding and good drivers.

10

The standard EQ for the TCF as I have seen on many ToT explanations is TCF = 650/(550+PHRF).
I have yet to find an explanation on the A or B factor and your explanation confuses me.
after recalculating the TCF using the standard Calculation and rescoring things look better.
I have reattached the file which has the PHRF scores the way we normally do our summer series manually changing the ratings. And the Alias are scored using the RYA NHC. But like Art pointed out with small fleets it does not work as well. Though there are some differences in the scores of individual races.

What is weird is I get a few error messages when the races are scored, and I cannot figure it out. Boats are entered but SW says they are not entered even though they are scored.
z RYA NHC 2022 WW.blw (153.9 KB)

Thanks

11

Dave,

Some follow-up thoughts.

  1. I didn’t realize you were already using skipper handicapping (aka
    personal handicapping). You have not explained why you think your
    current system is a failure and instead say you want to automate it. I
    am guessing you are unaware that Sailwave can automatically do the
    calculations for the system you are using. NHC with external mode takes
    the results of the current race, invokes an Excel spreadsheet with the
    formula you want to use and returns new ratings. No need to do it
    manually and no need to change from your current system of adjustments.

  2. The A and B factor numbers are average boatspeeds expressed in
    sec/mi. 6 kts equates to 600 sec/nm (1 hr / 6 nms = 3600/6 = 600).

  3. The A and B numbers you cite are outdated. Those recommendations were
    for a single rating system (So Cal PHRF has 3 ratings). And, they date
    back to a time when triangular courses (upwind, reach, reach) were the
    norm. I describe the methodology that you can use to generate a MUCH
    more accurate STD starting number. That derived STD is the “B” factor.

  4. To repeat what I said already, the A factor has no impact on the
    order of corrected times for boats. Mostly it determines which boat you
    want to have a TCF/ToT of 1.000. My recommendation would to use A = B so
    that a boat with a PHRF rating of 0 will have a TCF/ToT of 1.000. That
    will mean that a boat with a PHRF rating of 0 will have its corrected
    time = elapsed time under both ToT and ToD. That should make it easier
    for competitors to understand. And, it may give them more confidence
    that the ToT ratings, which they don’t understand, are fair.

Here is the logic of ToT PHRF ratings:

NOTE: ToD is only accurate to the extent you use an accurate distance.
Pre-GPS that was sometimes difficult and since ToT scoring doesn’t use
distance in the formula it could be more accurate. Many used it for that
reason. Plus, more importantly, in areas with heavy current (like high
tide areas or rivers) the distance sailed is usually much different from
the distance over the bottom, which results in significant inaccuracies.
Since the TCF/ToT scoring formula doesn’t use distance those
inaccuracies are eliminated. World wide you can see a correlation
between high current areas and the use of TCF/ToT ratings.

a. PHRF ratings are generally accurate for the range of average winds in
your area. It might be 8-12 or it might be 18-24 depending on your local
wind conditions. The point is you will tend to concentrate on getting
fair PHRF ratings for the wind conditions that are most typical.

b. When the windspeed varies from the average conditions for which PHRF
ratings are determined ToT is considered by mainstream naval architects
to be slightly more accurate then ToD, assuming the boats are all of a
similar type (NOT heavy displacement boats and light displacement
sportboats in the same race).

c. So, the goal of the conversion formula is to have the ratings/scoring
under ToT and ToD to be essentially identical in the average wind range
(since that is where they are considered accurate). You do that by using
a B factor that represents the speed (in sec/nm) of a zero-rated boat in
that average wind range. For races held in that average wind range the
ToT and ToD results should be basically the same (meaning order of
corrected times for the boats; of course, the actual corrected times
will be different). But, when the wind average is MUCH different (either
a lot lighter or a lot heavier) then the order of corrected times for
boats will be different under ToT than under ToD.

I think there is a lot of confusion about converting standard ToD PHRF
ratings into ToT ratings. Much of what is out there is plain wrong or
misleading. I’ve done a fair amount of analyzing PHRF ToD ratings vs ToT
ratings (easily several hundred races) and at one point the US Sailing
naval architect suggested that maybe I could write up an explanation
that they could post on the website. However, I never got around to
doing that write-up.

I would seriously consider using the type of skipper handicapping that
we used locally and I described. Imagine a 5-boat class where the
last-place boat corrected 2 seconds and the penultimate boat corrected 1
second behind the 3rd-place boat. For all practical purposes those two
boats were tied with the middle boat and the rating adjustment should
reflect that. At least, that is my opinion.

Good sailing,
Art