EHM The Blue Line

BTerran wrote:I definitely agree with this idea. It's a common problem in sports games, a maximum rating is handed out to star players when it really should be reserved for a once in a generation type of talent. A player like Gretzky should set the bar for what the maximum is, while a player like Sidney Crosby should have noticeably inferior ratings.

this is why attributes should be out of 50, not 20, or even 200 instead of 99, or something that will allow the league to get better and better, or create skill gaps between guys like Kovy, and guys that shouldnt be skating in the first place

Animal31 wrote:

BTerran wrote:I definitely agree with this idea. It's a common problem in sports games, a maximum rating is handed out to star players when it really should be reserved for a once in a generation type of talent. A player like Gretzky should set the bar for what the maximum is, while a player like Sidney Crosby should have noticeably inferior ratings.

this is why attributes should be out of 50, not 20, or even 200 instead of 99, or something that will allow the league to get better and better, or create skill gaps between guys like Kovy, and guys that shouldnt be skating in the first place

As far as I know,BGS' CM5 is using out of 100 max attributes,but unluckily,it's a bad attempt.so maybe out of 30,the game will be good-balanced

Good god...it does not matter what number is artificially assigned as the TOP RATING. It comes down to percentages boys and girls. If top number is 100 perfect player is 100. Average player would be 65. Barely capable 50 and useless below that. Ever take school percentages? Just relax about this. It is a RIDICULOUS conversation. The issue is how people rate players. Some think Scott Stevens was a perfect player, and would score him 100/100 or 20/20. Others think Wayne Gretzky is. Both of those are wrong but that is the way it is with fans. No player is perfect...and really ratings are just an approximation of WHOEVER is DOING THE WORK's OPINION on that player. Never going to be accurate exactly. No matter what you do.

Finally....if you want things to be more accurate and more to your own liking...then go ahead and do it yourself. If it works out I am sure most of us would love to see the perfection.

Except for the fact that averages increase

If 100 was the top anything in the league, chara would have 100 in shooting, and weber would have 99. Except last year, Weber shot as hard as chara did a year before, and chara shot even harder. What happens if that happens again? Weber shoots as hard as chara in the last all star game, and chara shoots even harder, then weber would have 101, and chara would have 102, then your "perfect player" dies. What happens in 10 years, when a guy comes in and breaks Chara's 2011 record by like 15 KPH? Not to say its possible, but under your system, his shooting would still only be 100

Thats why it should be out of something that allows for a skill gap, precision, and growth

and no

this isnt up to just one guy

everyone is allowed input, even you, thats why this forum is here. If you think its only up to Dabo, or me, then leave, and lock the forum, cause dicussion is usless at that point



in regards to inflation

An "average" player today, will not be as good as an "Average" player in 20 years. In 20 years, who's to say a guy like Kovalchuk isnt an average Russian Sniper, and a guy like Clarkson would be just good enough to play in the AHL, let alone get 30 goals in the NHL. It just wouldnt make sense to have 65 be average, no matter what, because 65 is 65, a 65 shot would be X MPH now, or in 20 years, but who's to say the average speed of an NHL slapshot doesnt because 1.5X?

Had a math lesson set up and typed. But there is no point. You can think the way you want as you say. I will continue to think the way I want. But math is math, and perecentages are percentages.

Animal31 wrote:An "average" player today, will not be as good as an "Average" player in 20 years. In 20 years, who's to say a guy like Kovalchuk isnt an average Russian Sniper, and a guy like Clarkson would be just good enough to play in the AHL, let alone get 30 goals in the NHL. It just wouldnt make sense to have 65 be average, no matter what, because 65 is 65, a 65 shot would be X MPH now, or in 20 years, but who's to say the average speed of an NHL slapshot doesnt because 1.5X?

True. But player ratings are not based on absolutes, they're relative to the known best at the time. 65 doesn't equate a given speed or something; it's in a way a percentage of the best know at the time (let's say Gretzky, for the sake of the example). So, today's average player with a skill at 65 means that his own ability at scoring or whatever else is "65% that of Gretzky" (well, maybe not 65%, but that's an example). In the future, another player might outshine Gretzky, but since the league average will also have increased, the average will be "65% of that new better guy" and no longer "65% of Gretzky".

Edit: post #500 Fine milestone.

Yup attributes are surely based on relatives and not absolutes. So a player with a 20 Slapshot has the best slapshot in the game - not necessarily that he can shoot at 100mph. A player with a 1 has the worst. If a player comes along with an even better slapshot, you'd give him a 20 and then reduce the attributes of some of the top players (but in reality, how likely is it that somebody will come along with a better slapshot than the existing top shooters?). With regards to the Slapshot skill, it's not just a measure of power, it's equally a measure of accuracy. Would you give a 20 Slapshot attribute to a player who has a 100mph shot but who always shoots wide? Or would you give a 20 Slapshot attribute to somebody who shoots at 50mph but who scores 1 in 3 of his slapshots?

Player attributes are really all about intangibles. You can't really measure a player's skill like you can measure a speed or distance. Hence ratings are relative to each other rather than absolute.

As for what range to use; it's all the same. Whether or not you use a 1-20 or a 1-100 scale, you'll end up with the same spread of skill. A player with a 19 attribute on a 1-20 scale is the equivalent of a 95 on a 1-100 scale. The only difference is the gradation. So a player rated at 96, 97, 98 or 99 would be the equivalent of a rating partway between 19 and 20 (i.e. 19.2, 19.4, 19.6 and 19.8). However, I think there comes a point where too fine a gradation makes things worse rather than better. On a 1-20 scale it's a relatively easy task to rank all of the players from worst to best. But when you're working with a 1-100 scale, surely it becomes harder - how can you tell whether a player is a 90 or a 91 - or a 92, 93 or 94?

P.s. Welcome to the half-century club, YZG!

Yes YZG congrats. And what youand archi said was exactly my point. Well said gentlemen. A term I use loosely when describing myself but not you two.

Except were not talking about later versions

we're talking about later in a single game

yes, in Dabo 2040, 20 in shooting might even be like the average shot now, or it might be 20miles faster, in which case, yes, we would scale

but in a single game, that will be played untill 2040, shots may get harder, or they may get slower. But what happens when the average decreases, and the top end might be impossibly high, in which case there wouldnt be numbers available for players, which is why numbers should be hard. We can easily call Chara a 20, and weber a 19.9. when a guy can shoot chara he can be a 21. Its much similar to allow for development, rather than program the game to change ratings based on the average of 15000+ players

Agreed -- but what Archi and I said still holds. In 2040, Chara and Webber will be long retired. The game will have generated new outstanding shooters, who will be given by the game the highest attribute values for that skill. And the average of the other players will have a lower attribute than them, as it will be relative to these new game-generated prominent shooters. Same for all other skills.

The thing is, for the game engine, having attributes superior to those of Chara or Webber for generated players when these are gone makes little sense, since when they are gone, there'll be no point for the game to compare the new players to them. If average shooting improves progressively over time, you can expect the other players' defensive abilities will progressively sharpen as well to counter it. Therefore, the better shooters of the time might not have a much greater effect then than now, given this. And if the average shooter got worse over time, then, out of 20, he'd reach 12 rather than 14, for instance, whereas the vastly superior elite would still fetch high attribute values. Even among newgens, attribute ratings remain relative. They tell us "this guy is this much better than this one at this thing" and helps calculate events in-game. No matter how good exactly the best shooter is, the others will always only be a percentage of this one guy.

I know you wonder what'd happen if the best shooter of all times started his career in 2040. The game wouldn't change the ratings of everyone because a top player spawns. That'd be a programming and gaming nightmare. It is simply assumed that all skills evolve at the same pace over time, so that the difference between the elite and the average is relatively constant. When the player spawns, by the time he develops into the very best shooter, others will have declined. And, the max attribute of 20 (or whatever scale is used) should almost never occur in game. The league's best player IMO should max out at 18-19. 20 should be for generational players only - that one guy you say outshoots Chara and Webber. So if a player reaches 20... he is among the all-time cream of the crop.

Also, remember what Archi rightly said. Shooting is more than just "firing it fast". Even if a guy shoots the puck at 150 MPH, if he always misses the net, he's useless. That attribute is more subtle to evaluate and conveys more information. Making "hard" number then becomes virtually impossible -- hence the need to rank players relative to each other rather than relative to a fixed scale.

P.S. Archi, half-century? That'd be 50 wouldn't it?

Half a millenium

unless the british are weird

For the purpose of this argument, im asuming the NHL 13 standard, of Slapshot power, Slapshot Accuracy, Wrist Shot Power, and Wrist Shot Accuracy, and

We wouldnt need to compare to chara when he retires, there would be other players that would shoot at "20" power, just like chara, we would compare to them. Hell, the current system is arbitrary as all hell, it should be absolute, with the power being a players average shooting speed in a relative vacume. We shouldnt base ratings off how they would play, as a player only has control over himself, its up to the defence to stop players from shooting 104 every single shot, thus nullifying the 21 shot power rating, no assumptions, none.

As a programming standpoint, its MUCH easier to calculate based on hard shooting values, ai aside. Instead of looking for this guys shot compared to every other, it would compare his 21, to the inputed 20 being 105-109, and say, this guy shoots 107, and has him shoot at 110, or 112, or even 120 based on what we chose to program, then compare that number to the situation, like how long does he have to get the proper weight transfer, is he shooting off balance, is the puch moving, etc

With absolute 20 being 108, and always 108, we can take the worst shot in the world, most likely 20mph [idk, what do pants 13 year olds shoot], or whatever, and say thats 1, then we can say a 91 is about a 17, and if a guy can shoot 150, then he can be a 30, not that its possible, but im not going to say that it isnt.

Same can be said of speed, and even accuracy, or anything. Well, accuracy should be percentage, you cant really be more accurate than 100%, so that would be how often a player can hit where he wants given perfect conditions. Speed would be how fast a guy can skate in a straight line on perfect ice, and full stamina. Game time, his actual speed would be based on his energy, room, etc

Everything should be rated on perfect conditions btw. I dont care what anyone says about this, it should be. A players competition should not determine his attributes. A guy with 20 power would only be affected by pants passes, or poor timing, or tight defence. To say "a player will not always be able to get his shot to full power every time, so he should be rated lower than perfect" is both arbitrary, and artificial, just like the "Scoring" slider. The amount of scoring in a league should be up to the sandbox, coaching, player styles, player skill, rink size, amount of penalties, not a slider defined by user

I dont want anything in this game to be hardcoded, or left up to averages, trends, or "user discretion". Everything should be possible. Japan should be a hockey super power if the right conditions are met. We cant assume that the average will increase with both min and max staying the same distance apart, that's artificial as well. the average isnt up to us, its chaos theory. Some leagues may get better, some may get worse. Juniors in Canada may become worse than juniors in Italy, with debatable possiblity

I like a good debate!

Personally I don't think the speed of a slapshot will change much over the years. Back in 1965 Bobby Hull set the record of a 118.3mph slapshot. How much has this changed over the past 48 years? Not much. Zdeno Chara holds the current NHL All Star record at 108.8mph, set last year. Granted, the accuracy of speed measureming equipment back in 1965 is questionable (apparently Hull managed a 96mph backhander...), but it's fair to assume that Hull's shot was in the region of 100mph. The only things that have changed over the past 48 years is the material in the stick and arguably the athleticism - but in fairness Hull was probably as much as an athlete as any modern day player (Hull used to be a farmhand). If Hull had a modern day stick, he'd probably shoot as hard as Chara. That's not bad considering he is a mere 5ft 10 compared to Chara at 6ft 9.

A couple of days ago Alexander Ryazantsev in the KHL All Star Skills Competition set a record of 114.127 mph. He's 6ft. However, it's not quite a fair comparison because the KHL uses a shorter distance in their Skills Competition than the NHL. Apparently this means that the speed reading in the KHL is more favourable - I guess because the longer distance to the measuring line in the NHL means that the puck may have already started to slow down.

I can't honestly see things changing much in the future. I bet Hull could have given Ryazantsev a run for his money if he had a modern stick.

As for rating shot speed and accuracy separately, I don't think this would work. Speed and accuracy are interlinked too much. Surely accuracy is relative to speed? So if a player has a 20 rating for both slapshot speed and accuracy, what does this mean? Would Chara's slapshot be at it's most accurate at 110mph? Or would it be more accurate at say 60mph? Would other players like Johnny Boychuk not be as accurate at that same speed? If in fact all players are at their most accurate at the same or similar speeds, how would this affect the shot speed rating? I think it's better to have an overall attribute measuring the player's slapshot quality rather than trying to break it down into its constituent parts. Ultimately, what we're looking to do is to simulate how frequently a player will be successful with a slapshot (bearing in mind that games such as EHM and Football Manager simulate the outcome and stats first and then simulates the action based upon the simulated outcome/stats).

Again, speed, stamina, etc etc are the same as slapshot. They're all relatives. You just rate the best player at 20 and the worst at 1. If you think there is the chance that a player will come along with a better speed/stamina then rate the current best at 19 or 18. This is how it's done in EHM, Football Manager, FHM and, as I understand it, OOTP (i.e. all of the top management sims).

Similar to my points about slapshot, think about how likely the speed of any future player will be faster than any current or previous player. In the absence of performance-enhancing drugs, probably not at all. In terms of professional 100m sprinting, we've probably reached a peak in terms of fitness. Sprinters are struggling to shave more than 1/100th of a second off of each others' time. In 1991 Carl Lewis set the world record 9.86 secs. 18 years later and the world record was broken by Usain Bolt at 9.58 - that's a difference of 0.28 seconds. How much faster is any professional sprinter likely to run? Probably very little. The reason being that we've arguably reached a peak in terms of sports fitness and nutrition. Without illegal performance-enhancing drugs, I don't see how professional sprinters will get any faster than maybe 1 or 2 hundredths of a second. If the likelihood of a professional sprinter getting any faster in the future is so marginal, what chance do hockey players stand when they're wearing pads and skates and carrying a stick on ice?

YZG wrote:P.S. Archi, half-century? That'd be 50 wouldn't it?

Haha oops! I clearly had dumb moment! 2 + 2 = 5, right?

I agree with archi and YZG,it is all about the gradation

But if we make the DHM the best thing,we can define each attribute/level to S(super),E(Excellent),G(Good),N(Normal),B(Bad),R(Rubbish) six level.eg,Pace E,Strength N,...,with different colors.But We definitely do not make a whole rate,it will make the game a piece of Adam Oates

This style let the game simple/nice/neat but more difficult

It will make the game more explorable,just like fog of war feature,the real attribue value determines the result of a match will be hidden at the database only and gamers can make the conlusion of roster. by plenty of match performance

We'd better make the data encrypted otherwise the scout tools will ruin it!!!

Do you all agree with me?

I dont want to do it like EHM, not at all, I want to do it better

And power and accuracy are defs separate in real hockey. I have a really good shot, and pretty accurate, not the best, but I can hit somewhere on the net :B But there are guys out there that can shoot harder than me, and pick corners whenever they want, somebody like kovlachuk. It seems like no matter how hard that guy shoots, it goes where he wants. so like I said, no asumptions, we cant asume anything. Some guys can shoot soft and still not hit the net, but can get it up fast, but still miss the net, so they would be 20/1, but in EHM they would have a slapshot as a 10, the game would interperate this as an average shot, not too fast, and not too slow, as well as hitting the mark alot more than a 1. Ratings should be all about precision, and allowing for any case in the real world, and for allowing something new. I know we all have EHM in our brains, but we should really think about how much EHM sucks compared to how it could be. It needs to be a sandbox in the purest form, no hard coding, no asumptions, no trends, no artificial anything, no possibilities unreachable

Stamina I think will be unreal in a few years. Compared to the old days, these guys are completly unhuman, in the best shape of anyone on the planet, and its only getting better, with guys like Gary Roberts being so strict in the fitness world, who's really to say players wont be more fit in 10 years


and yes, I agree that a players ratings should only be seen as they are scouted, unless its something like shot speed, or skating speed, as those can generally be seen just by watching a shift. Some abilities should be harder to see than others, like leadership, anticipation, stuff like that, and would be best left to "S, A, B, C, D,E, F"
where D is a whatever in whatever league you are in, with potentials to match. Those ratings should definitly be compared to other players

IE when a player has 21 in slapshot power, but everyone else has 19, it should just be an A, or even a B
unless players get their number value shown eventually with better scouts, then in that case slapshot should be the easier attribute to scout

So would a 21 correlate to what 150 mph who picks corners and never misses? Because that would be the perfect shot. Since perfect is and can only be measured as 100% the shot would be a goal each and every time. And 100% means u are setting the scale to 21 instead of say 20 or 100. And died would have 21/21 and all would be compared to that.

If u are setting the scale at 20 and giving some super dude a 21 you are over 100%. Beyond perfect. Not possible. Superman? Since computer programming and all designs for rating thjngs in sims are based on math --you cannot do better than 100% Amd the best shooter can only be rated the max scale.

In other words. It does not matter what he number for rating is set at...the best possible rating for any player would be tha max number. It does not make sense and would likely destroy the sim game to have someone rated at over 100% for any attribute. How much better than scoring on every shot you take can you be? You can't. Therefore a higher than 100% rating makes no sense unless you are playing a game in science fiction. Which is why EA ratings suck. Fire pucks guys. That is what arcade games not sports sims are for.

So of u like that type of game great. Enjoy it. Not in a sports sim though. They are mathematically based probability games. Anything else is outside reality and destroys the whole purpose of a sim game

Since Dabo is making a sim your though process Animal would not work. You Can not have a player shoot at more than 100% power. The top number you give has to be the top number possible. I get you could make other ratings in your shot theory lower so would not be perfect shooter and score every time.

Arcade games have their place but putting that kind of idea (super power burst) into a sim destroys the sim.

20 is not 100%

for accuracy it should be, in which case no, there wouldnt be a 21

but in no where EVER does it say 20 is max shot power, 100%, I said 20 is 108 MPH, not 100%, you can go faster than 108 if youre good enough
Its like you're not even reading. Even if 108 is 100%, in that case it is indeed possible to shoot faster than 100%

I am usin the number from EhM as example only. The max number you assign as maker of a game is arbitrary. No matter what number it is. Is still the max. Done with this because clearly math is not your strong suit. How in god will you code infinity.

It does not compute. I be what you are trying to say. You are not getting what we are trying to say. It's not me being mean or rude saying it. You jus do not get it.

In a SIM game. Sim. Game. Sim game. Not arcade game...I does not matter what the actual real life speed is. You are coding and the game translates who has a hard shot and who does not based on attribute setting. You have to set a max which would today be


Ah forget it...you are right. You have so validly proven that I will let it go.

bTw tomorrow u will arrive back on earth from my summer home on Pluto.

It doesn't happen in alot of NHL games though. Chara did it in perfect conditions and even there, it tooks alot of years to beat the past record. This isn't accurate IMO.

batdad wrote:I am usin the number from EhM as example only. The max number you assign as maker of a game is arbitrary. No matter what number it is. Is still the max. Done with this because clearly math is not your strong suit. How in god will you code infinity.

It does not compute. I be what you are trying to say. You are not getting what we are trying to say. It's not me being mean or rude saying it. You jus do not get it.

In a SIM game. Sim. Game. Sim game. Not arcade game...I does not matter what the actual real life speed is. You are coding and the game translates who has a hard shot and who does not based on attribute setting. You have to set a max which would today be


Ah forget it...you are right. You have so validly proven that I will let it go.

bTw tomorrow u will arrive back on earth from my summer home on Pluto.

Wow Lindros you too

I havnt defined the max number because we dont know what it is
20 is not the max number, as you seem to think
no, a player will never shoot about 100% accuracy, nor will they be able to shoot a number higher than the attribute, maybe 30

Dont say im bad with math, when you cant seem to think that 20 is 108mph, so a player should never be able to shoot above it

thats not how it works
and thats not what im saying
that has never been what im saying

Even in simulation, the game would say "this guy can shoot 108, this guy can shoot 100, this guy can shoot 110" and that will determine "this guy has a hard shot", not some arbitrary value, not what chara can shoot, but a number which is 5.4 times their attribute number

5.4*20 = 108mph
5.4*21 = 113mph
5.4*19 = 102
18 = 97
17 = 91

all the way down to 1 = 5.4

of course, we can change this 5.4 so the lightest shot in the game is 20mph in which case 108 would actually be 18 if you want to just add 3 to the attribute number, keeping the 5.4. Hell, numbers dont even matter right now, we could call attribute 20 300mph if we wanted to, it really doesnt make a difference. The entire argument in the first place was that players are going to shoot harder than chara, and will need an attribute range to say that he can shoot that hard without comparing him to another player in a 15000+ player database, or increasing the average arbitraraily to say that he is now the hardest shot in the game, but still only a 20. But I guess that means im an idiot and I suck at math, and I shouldnt be coding

great

Where does your 100% even come from? Chara? Was chara 100% 2 years ago? by your logic yes, was he 100% last year? No, he was better, Weber was 100%, chara was actually higher, so where does that leave the numbers? do we take everyone else down so chara is still 20? Do players get worse as chara gets faster? no, Chara has no bearing on anything, other than his correlation to the number 20, and the value of 108mph

a sim needs to take into account who's shooting, from where, how he's shooting, and how hard he's shooting, how many people are in front of the net to calculate the possibility of a deflection, how well a goalie can see, how big a goalie is, how fast a goalie is, where the goalie is, and even the handednesss of the goalie, and another ton of variables. Its not going to say "that was a hard shot, so that was a goal", that wouldnt be realistic.

Hell, im not saying that every player ever will shoot 20, but there will be guys, even chara, who will be able to shoot 21, or even 22, and should have the option open to be represented realisitcly. If its a sim, then why not have it be realistic?

philou21 wrote:It doesn't happen in alot of NHL games though. Chara did it in perfect conditions and even there, it tooks alot of years to beat the past record. This isn't accurate IMO.

Not the point, at least not to the extent you are thinking

the point is that chara would be able to shoot 108 in perfect conditions

the other team makes the conditions not perfect. If we went down, and said chara will be able to average 90 on any given night, then we will give him the value of 18, or 17 or whatever I just said above, I dont remember, then when the perfect conditions arrive, he will still only be able to shoot 90, because thats what we said

Its up to the coaches, and playmakers to set up chara so that he CAN shoot his 108, in the same way its up to the defence to STOP him from shooting 108, but at the end of the day, Chara can shoot 108, so his attributes should be reflective

Dear Animal

No need to be rude and call me a Lindros. And yes I can correlate a 20 or whatever rating you want to a 108 mph shot. But I don't care to continue this debate any further. There is not point. I already told you you must be right.

So young as well stop. You are a genius and the four people who told you you make no sense are weird, Lindroses, and will never understand math or how sim management games work as well as you.

I get that now. And I am not being facetious or mocking you. I am being totally honest and up front. Dabo should have you do all the programming and if he does not you should make your own game. I will be firs in line for it. It sounds like you have it all figured out and it should be awesome.

You have 4 or 5 people telling you it does not make sensein several totally irrational methods. But one of them now agrees with you. That is me. Archi Philou and YZG you just don't get it. This is the way math works. Chara has a 115 rated shot. Luongo has a 112 rated glove and 100 rated blocker. Plus Edler has 88 shot block and garrison a 95. Bergeron has a 55 deflection. Seguin a 99 passing. And all that means that when Chara shoots sometimes Bergeron will tip it. And sometimes it will go in. Other times Bergeron will tip it to Seguin who will have his pass blocked by Edler. Edler will pass to sedin. But sometimes luongo will stop the shot and Hamhuis will pass the puck. And so on. So yes the numbers that Animal wants to assign for attributes completely work. Do u Lindroses see it now? Clear as mud. Like 2+2. Simple math. 4.

Which by the way is the lowest rating of shot power which belongs to....wait for it....

Me.


But I scored on the rebound.

This arguement is so flawed my brain hurts.

Oh and BTW....it was just a joke like you posted in another thread. Take it as such. You are so very right.