rbodleyscott wrote:In what world view?Martin0112 wrote:
The the ranking will be different, and in this example, the difference is really important.
One where you need to get out more?
Moderators: philqw78, terrys, hammy, Slitherine Core, Field of Glory Moderators, Field of Glory Design

rbodleyscott wrote:In what world view?Martin0112 wrote:
The the ranking will be different, and in this example, the difference is really important.

Well please bear in mind that not everyone wants to use a computer program to score their tournaments, so the system has to be reasonably easy to use for manual calculation as well as by computer. A system with decimal points will never be as easy for manual calculation as a system without.Martin0112 wrote:But why making an easy calculating system (incl. decimals) worse by reducing it to full numbers?

While it is (usually) comparably easy to write a program that calculates with floating points and displays only results rounded to the nearest full number have you ever tried to get Excel to do that? (I can do it, but I bet 90% of the Excel users out there can not.) Given that Excelsheets are (sadly enough) still the most common form of computer aided scoring systems your argument about using floating point internally only is flawed.rbodleyscott wrote:Well please bear in mind that not everyone wants to use a computer program to score their tournaments, so the system has to be reasonably easy to use for manual calculation as well as by computer. A system with decimal points will never be as easy for manual calculation as a system without.Martin0112 wrote:But why making an easy calculating system (incl. decimals) worse by reducing it to full numbers?
Even when a computer is used, the players want to be able to easily manually calculate their scores because they (quite rightly) do not trust organisers never to make a data entry mistake. Organisers are only human.
And if we want to use the power of computers to the full we can make the computer use both integer and floating point and only use the floating point for tie-breaks. (Which means there is in fact no issue of 5 people in 5th place, since the computer can just as easily resolve tie-breaks for the lower places as the top three). Programming calculations is particularly easy, and the poor computer will not break or go on strike if it has to make calculations in multiple different ways.

When I first started playing FoG I wondered WTF! they were doing with the decimals appearing and disappearing and then a tie break on some invisible number nobody knew about. So I just stopped caring.Ghaznavid wrote:Using floating points only as an invisible tie-braker is not going to work IMO and frankly I don't see the evil of floating points or what's difficult with them. That might be cultural though. If one is used to measurements, etc. using the metric system he gets exposed notably more to floating points then someone accustomed to the imperial system (OTOH he is most likely less accustomed or comfortable with fraction stuff like 1/10, 1/8, etc.).

After each game. The scores for each round are based on rounded scores, so that they agree with the scores calculated manually from the score sheet without decimal points. The floating point representation is only used for tie breaks.Ghaznavid wrote:The question is also when do you round? After each round?
You are clearly missing my point. What I am advocating is that the official score calculation chart be changed so that it does not use decimals.Martin wrote:But, even the official score calculation chart is using decimals, so having the decimals noted instead of rounding results manually (also something to create errors) and add up the decimals is IMO as easy as adding rouded values.
It does not falsify the result. If it is stated in advance that the scores are based on the integer (rounded) scores for each game, then that is the rules of the tournament, so how can it be a false result?Martin wrote:Using decimals as a tie-breaker is what I can agree, but the real point here is: WHEN rounding the results!
As easily shown in my example on the page before, rounding after each round can (and in fact it does) falsify the result.


And I thought medicine was a science. Or are they still using leeches to cure people and blind faith to avoid driving off bridges in Usk?nikgaukroger wrote:And thus we run head first into the difference of approach between the "scientist" and those who primarily deal with people on a day to day basisAlways amuses me.
Medicine combines science and dealing with people. This provides us with a degree of omniscience and empathy only exceeded by the deity.philqw78 wrote:And I thought medicine was a science. Or are they still using leeches to cure people and blind faith to avoid driving off bridges in Usk?nikgaukroger wrote:And thus we run head first into the difference of approach between the "scientist" and those who primarily deal with people on a day to day basisAlways amuses me.

Surely you exceed the deity, as god does not think he is a doctor, but.......rbodleyscott wrote:Medicine combines science and dealing with people. This provides us with a degree of omniscience and empathy only exceeded by the deity.

You aren't earnestly arguing for using two separate sheets, do you? One of which either has to take it's values from the other or (IME more likely) people will enter the results twice (with the added risk of input errors).rbodleyscott wrote:After each game. The scores for each round are based on rounded scores, so that they agree with the scores calculated manually from the score sheet without decimal points. The floating point representation is only used for tie breaks.Ghaznavid wrote:The question is also when do you round? After each round?
I find the supposed difficulty with EXCEL difficult to believe. Surely it is just a matter of keeping a table of two separately calculated results for each game, then adding them up at the end. (The floating point version only being used to decide tie breaks in the integer version).
Not separate sheets but separate tables of integer and floating point results. And of course the data only needs to be entered once.Ghaznavid wrote:You aren't earnestly arguing for using two separate sheets, do you? One of which either has to take it's values from the other or (IME more likely) people will enter the results twice (with the added risk of input errors).rbodleyscott wrote:After each game. The scores for each round are based on rounded scores, so that they agree with the scores calculated manually from the score sheet without decimal points. The floating point representation is only used for tie breaks.Ghaznavid wrote:The question is also when do you round? After each round?
I find the supposed difficulty with EXCEL difficult to believe. Surely it is just a matter of keeping a table of two separately calculated results for each game, then adding them up at the end. (The floating point version only being used to decide tie breaks in the integer version).
This is simply not true. Perhaps you can e-mail me some time Karsten and I can explain better what I am getting at.IF one uses spreadsheet and IF one wants to use rounded scores and decimals for tie-breakers, the only clean solution is to create the sheet so that it calculates with floating points but displays only the rounded results.

We use the tournament program written by Thomas Peters. It does all the calculations and gives the results in the correct format for the rankings.nikgaukroger wrote: People are nasty, fuzzy and illogical - there is no logical reason why wargamers, who tend to be more intelligent than average, should have trouble with a scoresheet with decimal places in it, however, they do and do often. The best approach is to remove the problem.

Well, it's not that easy. As you might know, my tournament software allready includes the option not to use the floating point.rbodleyscott wrote:After each game. The scores for each round are based on rounded scores, so that they agree with the scores calculated manually from the score sheet without decimal points. The floating point representation is only used for tie breaks.
So, why not just doing a second score calculation sheet and offering both, giving the organizer of the tournament the choise?rbodleyscott wrote:You are clearly missing my point. What I am advocating is that the official score calculation chart be changed so that it does not use decimals.

We could do that, but honestly I don't believe much in the use of Spreadsheets for tournament scoring anyway, I was just mentioning them because they seem still widely used (although they often seem to me like part of the problem, not part of the solution). As Peter does in Italy, so far all tournaments in Germany (except the very 1st, which took place before the Software became available) did use the Program made by Thomas Peters* and hence never encountered any problems aside from late or (rarely) erroneous result slips. Since the result slips do not include floating points numbers, getting rid of floating points is not going to help with wrong result reporting.rbodleyscott wrote:This is simply not true. Perhaps you can e-mail me some time Karsten and I can explain better what I am getting at.IF one uses spreadsheet and IF one wants to use rounded scores and decimals for tie-breakers, the only clean solution is to create the sheet so that it calculates with floating points but displays only the rounded results.

Even if agreeing nearly 100% with what you saying, the problem you are mentioning here is already solved on the calculation table.KillingZoe wrote: ...
If you have a (not rounded) result of 22,5 to 2,5 (for excample) this will round up to 23 to 3, ruining the 25 to 0 score. So it is not sufficiant to round after each game, you have to round the loss and gain points of each player, then calculating the result by this.
...

If you feel like having a serious discussion on Saturday night after 3 games at BritCon you are much tougher then me.Martin0112 wrote: Maybe there will be a strong favorite in the future, let's see....
And I#m more than happy coming to Britcon this year to have a nice discussion at the evening or during the breaks.

Ghaznavid wrote: That said I still don't see the problem with results that have decimal places. Last time I was in the UK, neither the shops nor the gas stations used rounded prices, yet people seemed to handle that just fine (unlike me as I couldn't make heads or tails of those darn English coins).