The Friday Five: 5 Problems with Overall Ratings

Welcome to this week’s edition of The Friday Five! If this is your first time reading The Friday Five and you’re wondering what it’s all about, this is a feature that I post every Friday in which I pick a topic that’s related to basketball video games, the real NBA or another area of interest to our community, then either give my thoughts on the topic in five points or compile a Top 5 countdown.

Whether you’re simply playing basketball video games or creating roster updates for them, chances are that overall ratings have caused you some grief at one point or another. Since they are meant to represent a player’s overall value, skill level and to a certain extent their rank in the league, there is obviously a lot of interest in them being “correct”. However, that isn’t always a simple task and the significance of “perfect” overall ratings is often overblown. Therefore, today I’m drawing upon my experience as a roster maker and taking a look at five problems with overall ratings in basketball video games.

1. They’re very prominently displayed

Overall ratings receive a lot of scrutiny because they appear prominently wherever player ratings and statistics are displayed in-game. They offer a quick glance at everyone’s abilities and relative rank in the league, so finding out the top rated players in an upcoming game is something that fans always clamour to do during the preview season. Overall ratings have been in games for a long time – take a glance in the NBA Live 95 and NBA Live 96 PC editors if you don’t believe me – but ever since they started showing up in-game, fans have paid a lot of attention to them.

This invites some exaggeration of their importance. Overall ratings certainly aren’t irrelevant or useless but since they’re right there for everyone to see, a handy indicator of skill and value, we end up caring a little too much when a player is rated 76 overall and we believe he should be rated 78 overall instead. However, there’s usually a good reason that a player’s overall rating isn’t “ideal”.

2. They often aren’t calculated particularly well

When a game is calculating a player’s overall rating, certain individual ratings may not be weighted particularly well (if indeed they are weighted at all). This means that various factors are usually not taken into account, such as whether or not a particular skill is a significant strength or weakness given a player’s position and style of play. This makes it far easier to underrate and overrate players; more on that in a moment.

Fortunately, there is usually some wiggle room with individual ratings. The scale is large enough that you can sometimes fudge a rating here and there so that it’s still a reasonable representation of a player’s skills and will yield a realistic result in gameplay and simulation, but balances out the overall rating for aesthetic and ranking purposes. You don’t want to let the overall rating dictate the individual ratings too much as that’s not the right approach, but it’s possible to strike a good balance without compromising the accuracy of the individual ratings.

3. Some players are always going to appear overrated or underrated

There are limits on how much you can fudge individual ratings however and with overall ratings not always being calculated particularly well, some players are very difficult to rate accurately. Because Jumping and Dunking ratings tend to be weighted more heavily than they should be, players who are standout athletes and prolific dunkers often have a higher overall rating than they perhaps should, seemingly placing them on par with players who may have more all-around skill. Josh Smith is a good example of such a player, with his athletic prowess usually leading to a slightly inflated overall rating.

At the other end of the spectrum, you have players who are among the best in the league but have a few noteworthy flaws or weaknesses, which brings down their overall rating. If they also happen to be average athletes, their overall rating is bound to end up being at least a little lower than desired. Steve Nash is a prime example here; he’s been a perennial All-Star for much of his career, one of the best point guards in league history and is a future Hall of Famer. However, he doesn’t dunk, he’s far from being the most athletic player in the league and he’s not known for outstanding defense. As such, his overall rating has usually been a couple of points lower than it should be compared to other star point guards, but that’s difficult to reconcile without overrating him in specific areas.

When a player’s individual ratings are accurate, their overall rating is usually at least within an acceptable range. Until such time as they are calculated better, we just have to live with a couple of overall ratings here and there not being “perfect”; “close enough” has to suffice.

4. They aren’t as important as a lot of people think

Of course, it’s hard to sell a lot of basketball gamers on “close enough” when it comes to overall ratings. As I’ve already mentioned, they are prominently displayed and are a convenient measure of a player’s value. We want our basketball games to have realism and we want the players to accurately represent the skills and performance of their real life counterparts. Therefore, we also want accuracy in the attribute that’s intended to sum up how good a player is, compared to the rest of the league.

The thing is, when it comes to making a player play like his real life counterpart, his overall rating loses significance (except perhaps to suggest at a glance that he’s highly skilled in quite a few areas). Individual ratings and other attributes such as tendencies and Hot Spot data are the determining factors. Once again, if the individual ratings are accurate, then the overall rating is more often than not going to be reasonably accurate as well, give or take any quirks in the calculation. If Steve Nash’s ratings and attributes allow him to play and perform like the real Steve Nash, it shouldn’t matter if he’s an 84 overall, 88 overall or somewhere in between.

This is especially true of team overall ratings. If you’re still playing NBA Live 2005 and are using the NLSC rosters, you may have noticed that most teams are rated 99 overall. This is a by-product of the conversion process from the NBA Live 06 rosters, which I haven’t been able to resolve. However, bad teams are still weaker than elite teams and in simulation, the right teams are still on top of the league most of the time (the sim engine can yield weird results on occasion). That should give you some indication as to how much attention you should pay to the team overall ratings.

5. They are still important and factor into the experience

So, I’ve spent the first four points talking about how overall ratings often have overblown importance and probably shouldn’t receive as much attention as they do. However, having said that…you can’t get rid of them from basketball games and they do actually have an important role. That brings us to perhaps the biggest problem of all when it comes to overall ratings: for the reasons I’ve already mentioned, you can’t put too much stock in them, yet at the same time they must be paid attention.

First of all, you can’t just hide them because they do serve a useful function in representing a player’s skill and value. It’s handy to have that information, especially for gamers who are new to the NBA or are more casual fans. Second, if a player’s overall rating seems significantly higher or lower than it should be compared to the rest of the league, it’s a good indication that at least a few of their individual ratings are in need of an adjustment.

Most significantly however, because they are a calculation of a player’s overall value, they play a role when you’re offering trades to the CPU in modes such as Dynasty and Association. While the trading AI in most games from 2000 onwards will usually take into account other factors such as depth at a certain position, as I explain in this guide the CPU’s likelihood of accepting a trade proposal relies heavily on the value of the players it’s receiving, as represented by the overall rating. Putting together a trade proposal that’s likely to be accepted is therefore largely a matter of basic addition. As such, it’s important that we can see the players’ overall ratings, so that we are aware of their in-game value. When it comes down to it, we can’t get rid of overall ratings, so perhaps we all just need to understand them better.

That’s going to do it for this week’s Friday Five. I hope that you enjoyed it and that it shed a little light on a matter that often causes consternation for gamers and roster makers alike. If you’ve got an opinion or a question regarding overall ratings then have your say in the comments below and as always, feel free to take the discussion to the NLSC Forum. Please join me again next Friday for another Five.