Inquiry to experienced players and staff members about attribute range

Forgive me if this is a done to death topic. I honestly did search for some answers but didn't turn up much. I hope that both players and someone from SI sees this because they might have the best answer.

I recently ran across an argument on these forums where someone felt two of their favorite players weren't being given a fair shake in terms of attributes on Day 1 of a new game. They felt that 1) their preferred players were a bit too low and 2) their biggest rival's players were way too overblown. Cue typical "researchers are biased" argument. OP of that eventually got banned lol.

Anyway, in the thread, someone from SI (forgot who sorry) made a really good point. He stated that the difference between a single point in a given attribute is very important. He suggested that this person think in the scope of the game where there are tons of leagues consisting of sometimes very low quality (in comparison to the big leagues.) He suggested that based on OP's logic, people in leagues like championship would be stuck in the 11-13 range and BSN/S people would be stuck in 7-10 range while the lowest leagues from the most obscure countries would be stuck in the 2-7 range for all attributes (in order to maintain comparison across the game.)

My question is this: Why was the range of 20 chosen? Besides the answer of "that's just how the game works," what was it that lead to such a small range? Its clear that there is a far bigger difference in class between Ronaldo and a player from the Blue Square North than the range of 20 can provide. Of course, it would require massive engine changes to come up with a new scale of 50 or 100. But why not just double it? 0 to 40 seems like it would solve the range/accuracy issues while not requiring much change in terms of the math and engine.

Thanks.

The 0-20 attribute has always been in the game since the CM days.

In those days there wasn't the amount of research and leagues available that there is now so it probably more simple to have that range instead of up 100.

Now they would have to re-evaluate all the players to give more precise ratings and that wouldn't stop the argument you raised about my player should be better than player x in a rival team.

Also they would have to change the data input ranges of the match engine which would affect the calculations and the ranges in that.

Are there not two decimal points used in attributes? So that the range is actually much bigger than we see? That's what I'd assumed anyway.

The game does operate on a 100 point scale, so a player with Passing 14, for example, actually has a Passing attribute of somewhere between 66 and 70. If you look at the progress charts on the training screen, you can get a more precise sense of your players' actual attribute rating.

Accordingly, I assume that this means that a difference of one point in Passing with all other things equal represents that the superior player has a 1% to 9% better chance of making a pass than the marginally inferior player, so a difference of one point can either indicate a fairly big difference or no real difference at all.

As for why a smaller range was chosen to represent what is actually a larger range, I suspect that is probably meant to convey a bit of ambiguity about the precise ability of a player (especially given that many feel that a quantified attribute system in general is unrealistic and should be replaced wholly with scouting reports that use relative, qualitative terms).

Are there not two decimal points used in attributes? So that the range is actually much bigger than we see? That's what I'd assumed anyway.

This, As I understand it the ME works from 0.1 to 20.0 giving a range of 200 rather than 20.

In terms of what we see researchers I suspect only rank from 1-20 and given the arguments this causes it would be worse asking them to rank on a 200 scale.

This, As I understand it the ME works from 0.1 to 20.0 giving a range of 200 rather than 20.

In terms of what we see researchers I suspect only rank from 1-20 and given the arguments this causes it would be worse asking them to rank on a 200 scale.

thank heavens for the 20 point scale

i can actually imagine the chaos that would follow if it got changed to 200

*why has Rooney only got 192 for Determination, he should be 193 imo*

Thanks everyone I appreciate the info. I didn't know about the decimal system but now my training line graphs make more sense. That definitely does bring a bigger range than I thought. Also, the point about the "player x is better than y" was just a lead in. I'm sure nothing could satisfy that argument.

I do agree with samdiatmh and think a 200 point scale would be way too much. However, I still think that even with a decimal system that has a big range, that range isn't expressed. Its easy to use an extreme example like Messi vs. a South African league player but I do think the system is limited. For example, a striker with an off the ball of 4 would be worthless right? As far as I know (I don't play much LLM) a 4 is a terrible attribute. But if you compare an average to lower end player in the Turkish Super League versus a middle to low end guy from South Africa, there has to be a finer difference than 1 to 20 (even if there are decimals because they don't matter for us so much, just the ME.)

How would people feel about a 40 point scale? The math (since its just doubled) would be easier. I know it'd be harder for researchers but I'm only asking hypothetically.

Thanks everyone I appreciate the info. I didn't know about the decimal system but now my training line graphs make more sense. That definitely does bring a bigger range than I thought. Also, the point about the "player x is better than y" was just a lead in. I'm sure nothing could satisfy that argument.

I do agree with samdiatmh and think a 200 point scale would be way too much. However, I still think that even with a decimal system that has a big range, that range isn't expressed. Its easy to use an extreme example like Messi vs. a South African league player but I do think the system is limited. For example, a striker with an off the ball of 4 would be worthless right? As far as I know (I don't play much LLM) a 4 is a terrible attribute. But if you compare an average to lower end player in the Turkish Super League versus a middle to low end guy from South Africa, there has to be a finer difference than 1 to 20 (even if there are decimals because they don't matter for us so much, just the ME.)

How would people feel about a 40 point scale? The math (since its just doubled) would be easier. I know it'd be harder for researchers but I'm only asking hypothetically.

Whilst I understand what you're saying, what you're arguing for seems to me to be entirely cosmetic. The 1-20 scale works pretty well in representing the difference between players to us and in translating different abilities into the match engine. The only time it perhaps breaks down is with once in a generation players like Messi, but even there his performances in the game are generally not far off his stats in real life. Also, if the scale were to be increased it would probably cause problems when representing games between a top club and minnows - if the difference were too large they could easily turn into cricket scores. Perhaps this could be overcome by boosts to morale and motivation, but it becomes trickier to balance.

I'd also suggest that the difference between, say, a Premier League player and a Blue Square Bet Premier player is not as great as you might imagine. It's easy to think that the average Bath or Southport player is a no-hoper but they're actually very good footballers - far far better than most of us who played at school or have a kickabout in 5-a-side. What separates the top players from semi-professionals is generally the abilities represented by the mental and hidden attributes: reading of the game, decision-making, tactical awareness, personality etc. Is James Milner ten or even five times better at passing technique than the average Southport player? Possibly, but I'm not convinced.