Playing "stats only" in text sims
 

I have begun to play OOTP stats only, turning off all ratings and it has made the game much more immersive to me. I was so hung up on looking for 8's, 9's and 10's in different categories, that I lost track of actually looking at stats to see if a player was a fit for my team. For me (and I know this doesn't go for everyone), pouring over all the stats for each player to see if he's a good fit or if he's for real helps me connect to my league much more than ratings.

Granted baseball takes place in 9 vacuums so it is much easier to figure out if a player is good or bad based on stats.

I am thinking of doing stats only in Fast Break, but am concerned it will be much too difficult and I definitely don't know the metrics as well in basketball.

Anybody play stats only in their games?

For those that don't do you think this it could be a viable/fun option in baseball, football, basketball sims?

About 20 years ago, there was a lot of focus on MLEs (Minor League Equivalencies) in sabermetrics. The idea was that you could create a set of translation functions that could be used to evaluate potential major leaguers.

Taking the scout out of the equation is very appealing, but unfortunately, development curves vary considerably and MLEs never proved all that reliable.

With football, it's nearly impossible, even from college to the pros. Imagine trying to project what Danny Woodhead would be as a pro based on college numbers alone.

In terms of pure number crunching, a good simulation could create a minor league world or a college world that would translate well, numerically, by statistic alone. That might be more fun than ratings from a gaming perspective.

But, really, in terms of simming, your frustrations are more that developers haven't come up with a good way of accurately and in an entertaining manner, capturing how scouting works.

I will sometimes play OOTP this way. I agree it's a pretty immersive way to play. You can't play quickly that way, but it's a lot of fun when you have time to kill and like a bit of fog of war and want to play with just stats and nothing else to evaluate guys and make deals and stuff.

One of the many things I miss about my favorite horse racing sim was that it didn't give you any ratings at all. You had to figure out what class your horse was and what distance it preferred by workouts and races.

Really had an effect of both increasing the challenge of the game plus helped with the immersion -- instead of min/maxing ratings I was forced to think along the lines of "X faded a bit after an easy lead at 1 1/8 miles but that was against 40K claimers.....should I try him the same distance against lower or shorten him up a bit?"

I've always had trouble getting immersed with OOTP. Maybe I'll give this a shot and see if it helps.

That's an interesting thought, Path, that the scouting or the practices could become a part of the football sim. I often want to try something out with a team but don't care experiment in the regular season games. But what if we could experiment in practice and get some sort of results that could hint at both player ability, and how well a new gameplan might work. That would be interesting.

Hope it helps for you on OOTP!

Your thoughts on the horse racing sim is how i envision it for football.

Take the draft for instance:
Combine numbers are given for each player along with their college stats. Each college player has a game log so now you can see how they accumulated their stats. So for example perhaps a DE from the SEC has 20 sacks but 12 of them came against bottom feeder competition. The numbers are good, but they are skewed against poor competition. Do you roll the dice?

Or a WR with great combine numbers but comes from a program that runs the ball a lot so he doesnt have great stats.

So if either player booms or busts, there is no formula to it or it doesn't feel cheap. There are logical explanations for both outcomes.

I think the reason I don't like scouting is because in text sim games I am the owner, I am the gm, I am the coach. I have a certain amount of direct control. Scouting feels like a cheap way to throw fog of war in the name of realism or challenge. If I am given just the stats then I can be the scout and interpret the numbers , which to me feels more immersive.

But I do understand the massive challenge it would be to implement. Why did Kevin Kolb bust when he switched teams or why did Alvin Harper back in the day? Without ratings as a basis I can understand how that would very hard to simulate in a stats only environment and for many people it may not be an appealing way to play.

I'd like to give this a try at some point in OOTP - what is the easiest way to disable all the ratings? I can probably find it but haven't checked.

League Setup->Global Setup->Rating Scales

I've tempted myself with this before. OOTP also has "talent only" in older versions (at least) to give you limited scouting assistance, while leaving the judgment of actual ability to you through a player's stats.

This. The problem with only stats is that that approach isn't any more "real" than the OOTP ratings. Personally, I think the problem is the ratings and scouting reports are too few, but adding the thousands of possibilities that I think it would take would be nearly impossible.

In a perfect world I'd like to have a discussion with a AI scout and have to drill down through initial numbers to get more insight and comparable players. I'm not sure a whole lot of players want that that kind of time intensive process though.

I think replacing number or letter ratings with measurables plus a paragraph detailing the players positive and negative traits would be a good way to replace ratings. To add ambiguity, the outlining of the players traits can use multiple different descriptions. Good hands and soft hands and great hands and hands like glue could all mean roughly the same thing. using that, it would be up to the user to determine through observation whether or not those statements are true. I think it would require a lot more information to be delivered in the form of statistics. We would have be able to pick apart players down to their individual match ups.

I think that would be cool at first, but very easy to figure out. If five or six phrases mean, good kicking distance or good route running, then when I see them, i see ratings. Yes, they might be wrong, just like ratings might be wrong, but it would just be verbal code for ratings. It would be difficult to keep a text and stats only scenario that would keep things fresh season after season, after season.

There was a soccer game that tried this many years ago, Sick as a Parrot maybe? It actually had no rankings, just descriptions.

I really like HR's method for the Fast Break series: exact current ratings (which to me include all the things you can't easily represent, and you generally should know how good a player is right now based on practices and watching game film, what his strengths/weaknesses are), with a vague guess at the future ratings (letter grades) in each area so you have some idea of what the growth potential for a player is.

ping Mac Howard

This is why I would always end up abandoning the career when I tried to play stats-only and have come to the conclusion that it's not for me. I want to play multiple seasons in one setting, not get mired down in trying to evaluate players. I use the most granular ratings scale possible in games like FBCB and OOTP that offer options for which ratings scale to use for this reason, too. I've realized that I'm just not interested in poring over stats to figure out which 6-contact player is better. I'd rather just sign the 62 guy, cut the 56 guy, and have a tight or self-imposed salary setup that gives me less room with which to work.

It makes me wonder whether there's documentation out there that reveals how scouts work in the NFL (or in other sports as far as that matters to the topic).

If the movie "Moneyball" is any indication, they base everything off the attractiveness of the player's girlfriend.

Sounds great in theory, isn't practical. Do we really want to read a scouting report every time we're looking at some guy? Do you know how many fictional players fill up a league?!

Its nothing you wouldn't see on tape. If a guy has good hands, he has good hands. Does it translate? Does the sum of the parts increase the value of the whole? Gotta play him to find out.

John Doe has solid hands and a large catch radius, does well boxing out defenders with his size and good body position. Has trouble with separation, but is a technically efficient route runner who attacks the ball in the air. Fights for yards after the catch, but sometimes tries too hard resulting in minor fumble issues. Has potential to be a solid possession receiver in the right system, but work ethic and overall intelligence has been questioned.

Just like a real scouting report, it is an opinion that can be completely off base, say a range of 50%. You would have to put the guy out there and then see how he does, no matter how long you played the game. There were people who thought Vernon Gholston was going to be a megastud.

Then you see deep statistical representation of his on field exploits. How many plays did he play? What percentage of those plays did he run a route? What routes did he run? How often was he open according to the tape? How often was he targeted? Was it against man or zone? Was he double teamed? When he blocked on running plays, what was his grade on those plays? Who was he running routes against or blocking? Did he tend to win against certain corners or safeties? Was he better at finding holes in a zone or beating his man one on one?

I think the only way to have something like this without having a visual representation of the happenings on the field would be to have an insanely deep sim engine that handled individual matchups on every single play. You would need to know your opponents players to some degree in order to make an educated guess as to how well your player did.. And what if he is just in a slump? What if he is just lacking confidence? What if its the coaches influence on him that is hurting his development?

EDIT: I do whole heartedly agree this would be far too tedious for most. I also believe its the only real way to do football "ratings free". In terms of FOF, I think the best way to implement something like this would be to make scout ratings WAY more erratic. You would also have to eliminate the direct ties that combine numbers have to do with a players skill set. There are some players who are big play receivers who don't run a 4.3. Case in point being Jerry Rice. One of the biggest big play receivers ever. Ran a 4.7. a 4.7 receiver would NEVER be drafted in FOF for his big play ability. It needs to be possible to find guys who are great at things they don't appear to be. Not only via volatility, but also via development. With development tracking being introduced into the game, maybe there is an opportunity to make it way more random and less cut and dried.

i thought stats only or scouting report was a cool idea til i got 2 lines into that paragraph and my eyes glazed over. if i had to read something like that for every player, i'd quit the game.

It's been a while since I played FB but I like the way this sounds. Personally, I got kind of turned off by FOF when I felt like I needed to start "solving" what the ratings bars likely meant. It felt like cheating to me.

I say eliminate the green bars, give us only combines and college stats and injury history to go off of (maybe with a rating for "quality of competition") and don't reveal rookie ratings until after camp. Then just give us red bars that will go up or down or both from week to week (mostly based on confidence and health), and to a greater extent from year to year (mostly based on potential and work ethic).

i actually really like the idea of eliminating green bars, especially since a change tracker will be incorporated into fof7

That's true, I didn;t think about that. Nice point.


EDIT - I believe you have swayed me!

This is what I was saying above. I like to think more like a CEO than to do grunt work. Just give me some numbers and let me make a decision based on those.

For OOTP, I don't read the scouting reports. I go by the numbers as well. They just happen to be stats, not ratings.

Perhaps Jim can add a feature to eliminate all the ratings without adding any superfluous scouting reports and see if people take to it.

The one thing I'd wonder about that is the fact that people are SO conditioned in FOF to the draft being far more predictable than real life. I cranked up volatility on all players to 100 in the BFL and some people absolutely freak out about it, despite the fact that it means that it only doubles the number of guys who bust--meaning that the average first round in the BFL has 2 busts instead of one--wayyyyyyyyyyyyyyyyy lower than the NFL. I think a lot of people *say* they want the draft to be less predictable, but I'm not sure that they'd actually enjoy it being less predictable.

Hard to prove you're better than 31 other guys at gaming the system if there's more noise thrown in ;)

SI

I for one would be all for a WAY more volatile player landscape. I want to see guys go up and down drastically compared to the current way it is done. Small/medium movements from week to week to simulate streaking/slumping, and way more significant changes during the offseason. More volatile guys routinely jumping 10 points on some categories while dropping 10 in others, more frequent suspensions for red flag guys.. I'd love to see some occasional in season suspensions for repeated unsportsmanlike conduct in games or fighting during practice or breaking team rules for those red flag guys. I'd like to see more red flag guys in general. Flags for guys with questionable motors, off field issues, locker room problems, personal foul risks, selfishness, diva behavior, weight issues, multiple problems on some guys.. In the same way, I would love to see some green flag guys. Good character, locker room leaders, workout warriors, students of the game.. I'd like to see some guys with green AND red flags.

Maybe also give us the chance to greatly influence how a player develops by having the coaches focus on specific aspects of their game in the offseason.

Okay Julio, now you are moving into territory I discussed on here like ten years ago ,when I suggested that the sim add traits to characters. I had a big list of sample traits, like "Good under pressure", "Good on Evening Games," "Fades in Postseason," "and a bunch more. From a kicker who's balls only were affected half as much in bad weather to a defensive player with a heightened ability to scoop up a loose fumble, these skills would be on a variety of players, and make people more than just stats.

I don;t know if anyone has that list around anymore that I dreamed up, but you get the idea.

Yeah, I think making players more volatile by adding these types of traits to them would make it easier for people to accept when they make big jumps or big drops in talent from year to year. At least then it wouldn't seem totally random. It would be nice if a red flag guy could also lose his flag under the tutelage of a good mentor and vice versa. Maybe a green flag guy on a directionless team becomes disenfranchised and pulls something like randy moss. If traded or let go, maybe he returns to his old self at his new home, maybe he doesn't. maybe he develops a permanent red flag from that.

This is where the choice of game vs. simulation becomes more paramount than most people understand.

Are you going to go the 'make an intriguing game' route that Abe suggests?

Or the 'make the most detailed, accurate simulation of reality possible' route?

If you make a game with say 'Clutch ratings' then the Sabremetric statheads will go ballistic with studies that show there is no such thing.

If go realism, like the many current suggestions over at OOTP for additional staff and scouts, then you risk making a huge, slow and tedious BUT ACCURATE (arguably) monster that gamers shun.

So I continue to think that a designer should state upfront as much as possible which items will be game and which will be simulation in an attempt to control and direct comments along his development plan.

Not that this will stem criticism, but allow a designer to point to his intentions and say "Look, I simply chose not to go down that programming path upfront. Possibly someone else could make a game/sim with that feature, you, perhaps?"

Hrm, I was thinking about MicroLeague vs OOTP last night, and I guess they fall along those lines (or if you want to sub in something like PureSim). MicroLeague - you input stats and the game tries to approximate those. More of a sim. OOTP? Ratings-based. More of a game, really.