Premiership Coach

[random260]

You're a legend Grenville.

My footy game is going pretty well. I haven't actually ended up taking anything from PC2011 as it turns out (got about as far as getting it to compile then got busy and never went back), and I think that's appropriate because I think they're going to be fundamentally different experiences anyway. But this is the Premiership Coach thread so that's the last time I'll mention this.

So now I'm going to shift back into "fan of PCXX" mode.

I have STILL got my original save file that I started when I first played PC2011 and it's still going strong. Everytime I go back on a PC2011 binge every 6 or so months, I just play this save. Gold Coast are in 2094 now and have won 52 premierships. I have played every single game as real time match day. I'm really not sure if this is the oldest consistent save or not. I think it wouldn't be too difficult to sim all the way to 2094, but the distinguishing thing about this save is that I have lovingly played every single moment of it. Not a single season was simmed. Not a single game was skipped. Not a single team was assistant picked. Not a single draft was recruiter selected. Not a single contract was AI managed.

At the end of this, the main thing that I feel the game lacks is meaningful statistics. Top 10 lists just don't cut it for me. But for this, I've gone and made my own and that suits me well enough.

However, I'm going to try and raise some meaningful points about the state of the game that I play here in 2094.

1. I do not win every game, and certainly not easily. I won about 17 premierships in a row at one stage, and barely missed a grand final up until around the 2050s where the clubs all somewhat balanced out. While this took a while, it's frankly a whole other game when you get here.
2. Staff were the first to break. This is going to be the first of a few of these similar sounding issues. I simple cannot offer anyone enough money. Even the duds. The input just doesn't let me. For a while, I could get around this by letting the AI sign my staff. However, now, even the AI has given up trying to fit the staff expected salary into my weekly budget. Since the staff have all unanimously rejected my club, this means I no longer get the rich recruiting experience. To be honest, this wasn't too bad. I had to pick players off of their position, height, weight, age and agility, etc scores. I also use a "trick" which I'm not sure if it really works, where I go to the F/S area and make sure the list is ordered by potential descending. I've somewhat purposefully never tried to figure out if this does work or not. This whole point may have been a reason why the competition got stronger, but I don't think so, because I still did wonders on the trade table and just focused on ready made players more than youth.
3. Finances took a turn for the horrible. It pains me to say, but I regularly edit the .xml file of my club just so that I don't have the worst facilities imaginable. I cannot get the weekly allowance to stick, because the club just won't let me turn a negative weekly balance. I haven't got any staff. I have got over $400million in reserves but I can't pay within barely 30% on my facilities as is required to keep them level. The inputs let me put in the right values, but every week the game changes them for me back to small amounts. It's as if the club just won't let me lose money. As a result, if I get lazy, I get to watch those overalls decrease, and that's painful. Every other club in the game has A+ everything. However, I think this plays a major role in the equalisation of the comp. So it's a blessing and a curse.
4. Player contracts are no longer micro manageable. It's as simple as I simply cannot offer any player a contract worth more than $50million. So, over 5 years, that's $10mil a season. My good players are expecting upwards of $12mil. I have absolutely no trouble fitting them into the salary cap. But I just can't sign them for an extended amount of years anymore. The only way I can sign these players is by accepting what they want, which is working fine for me so far, but is definitely going to contribute to equilisation in the future, because...
5. In my opinion, the best way to succeed in this game is to front/back load contracts. I regularly end a season with my weakest 22 player 82+ rated, at least a couple 90s, 31 players signed for the following season and 70% of the salary cap free for that off season. This goes in tandem with...
6. Trading. Back loaded contract? Trade him for the last year, where 100% of the contract was loaded. Can't get a club to agree? Give them a draft pick and take all their aging verterans. All you have to do is check that their contracts are expiring, and simply don't renew them. What's better is you can sometimes even offload them individually to other clubs to get back more draft picks than you lost. All in all, you got a player for free for four seasons and ended up swapping him for draft picks. Granted, the picks aren't normally very good (40s) but this works because you only do this when the player is ~33yo in the last year of their contract. However, this isn't even the best way to get an advantage out of trading, because that's left for...
7. Signing as many preseason drafts as you have picks for. Since I've got so much cap space, I can use every single one of my preseason picks even if I've already got 39 players drafted. I think if I have 40 it doesn't let me, but it only seems to check at the start of the draft. Most other clubs only pick once if at all, but normally there are still a bunch of Excellent or at least Good rated players still available. The kicker, though, is that a lot of the time these are players I traded away for draft picks during the trade period that had just passed. It's scandalous.
   
   But to be honest, I did away with a lot of these techniques after trying them for a while because it's just not any fun if you're effectively cheating. I front load contracts only now, so I pay for the 5 years up front and then get 4 free years. This does wonders for building a salary cap advantage and keeping it. I also always make sure I use every bit of cap space, sometimes restructuring contracts that are set to expire the following season. This sometimes results in paying more overall, but lets me keep the advantage.
   
   
8. The match stats are all dead. Once every few seasons, a player from an opposition club will have a remarkable game where they did something like kick 17 goals. This is a great accomplishment on its own. But the game goes a step further and wants that accomplishment remembered not once, but up to 10 times, where week by week the accomplishment counts again as if it happened again. Really, it happened once, but now the top 10 goals kicked are 19 goals by Player A, 18 goals by Player B, 17 goals by Player C, 17 goals by Player C, 17 goals by Player C, 17 goals by Player C, 17 goals by Player C, 17 goals by Player C... you get the idea. This has happened to practically all of the match records, except now when I notice it happening I immediately save and quit so it stops, and I've manually cleared my match records to resolve some really ugly ones. It even happened to my own player once! He got 71 possessions, and that was outstanding, but he only did it once, despite what the game wants to believe!
9. Quirks of the real time match engine. I'm going to brush over this quickly. But in summary, to be honest, this is actually really charming and I kind of love it. It produces some really interesting scenarios sometimes and some feel good stories.
   1. Inside midfielders are the best LFPs the game has ever seen. Regularly kicking bags of 7. I had a player called E.Nother who kicked 1100 goals and had a Small and Tall Forward rating well below noticeable, but he was a 88 inside mid and practically nothing else. I loved imaging the commentators explaining "He's got E.Nother!". Another player of mine, similar, kicked 133 goals in a season as a 84 inside mid and not much else. Once I discovered this, instead of trading them I started keeping these sort of otherwise unversatile players as if they were my next Eddie Betts. It "feels" right too.
      
   2. Small defenders are the best forward flankers / wingers, and they're rare. It is so hard to find a good small defender rated over 75. I'm pretty sure I always have them all. But sometimes you find gems that are 84 rated small defenders, and good inside, and that's when you put them on the wing. 700+ possessions season after season.
   3. Tall forwards are the best midfielders. This is so much better than the small defenders, and much more exciting. They're also tricky to find, but S. Stuthridge is a good example. At his peak, he was an 84 in Small Forward, Tall Forward, Inside and Outside Mid. This made him an absolute ball magnet. He was the player to get 71 possessions, and in that game, he also kicked 6 goals on a forward flank. Later in his career he played rover as a 34yo, rated 74 in his highest rating and still knew how to collect 28 possessions a game. Freak.
   4. Opposision tall forwards can get on hot streaks and that's what kills you. When an opponent's key forward kicks 3-4 goals in the first quarter, I know I'm in for a management challenge. I can't swap anyone onto him to stop him. He will end up with 15 goals and I will struggle to win. It's electric.
10. Pretty sure I have never, ever actually had a player who was "called" a Small Forward. Do not rate them at all.

Anyway an hour later this is too much to keep going now I'm going to stop here. Great game Grenville and I still love every second of it.

I'm in year 2125. Previously I was able to get AI to sign up assistant coaches but ofc it doesn't hire due to this issue.

At the current stage when I go to AI hire assistant coaches it now crashes.

How do you resolve this issue.

The only workaround is potentially manually signing assistant coaches to all 18 clubs. But even then idk if that will bypass the issue.

 

[Simon_Nesbit]

I'm in year 2125. Previously I was able to get AI to sign up assistant coaches but ofc it doesn't hire due to this issue.

At the current stage when I go to AI hire assistant coaches it now crashes.

How do you resolve this issue.

The only workaround is potentially manually signing assistant coaches to all 18 clubs. But even then idk if that will bypass the issue.

Haven't seen this thread for a few years - nice to stumble upon it and find somethings happening again.

2125 is an excellent effort, I usually start running into errors around 2050/2060.

Regarding the assistant coaches, I found the best thing was to edit/extend their contract end dates once they got ridiculous - I used a txt editor app to bulk change (ie "contractendYear>2025" change to "contractendYear>2026". I remember there was a lot of fiddling around to record the right line (I think the generic "Year>" was used a couple of times in each file).

I did the same with "marquee" players (those on max $$$ contracts) - if unsigned after Computer resign day they got an automatic 1yr extension. Although never played long enough than to have a couple hit this mark.

You can edit the clubs cash reserves (I was a little confusing but one of the $$$ figures under clubs does affect signing behaviour) - in fact this was the first balancing issue I found, with a couple of clubs billionaires, whilst others were virtually bankrupt. Some form of revenue sharing would be handy (or indeed simply turning off the artificial inflation).

The biggest thing I wanted to get fixed was around player generation method and the gameplay engine - reducing focus on KPF and Ruck; significantly increase KPD/SD values; and improve selection algorithms to consider secondary positions (instead of picking a 65 Small Defender over your 9th midfielder who is 75 IM/74 OM/73 SD). Ad was working on some changes but such a huge adjustment meant changing the entire gameplay model, so he was holding out for the new version (which never came. )

With new player generation (Regional) - the game 'brute force' created players based on highest skill point - which rewarded 'min/max' players. Anyone who played often with the original 2010 lists observed the dominance of Hawthorn 2011-2016 - as I crafted each Hawthorn player by hand - they were all 'rounded' players who were at least average in other roles - KPD could also run and kick, BP could mark and win the loose ball; midfielders could mark and forwards could defend. I had tried to overcome the KPF issue, but it was too powerful, and the first few drafts kept bringing in new dominant KPF who kick 1000+ goals anyway.

I managed to generate more reasonable stats with the engine in my solo games, but this required significant modification to all clubs player ratings every year (neutering anyone who even looked like a KPF, boosting defenders until they were almost as highly rated IM/OM).

It's also pretty easy to setup a "Zone" or "Academy" alignment using the existing father-son bidding engine, but found clubs massively overpay for mediocre/poor draftees - again for my solo games I'd remove the FS qualification for any that simply weren't going to be a 1st round pick calibre player.

I haven't got the game anymore, but happy to answer any questions I can for those that like to 'tinker' with the ratings/csv files (no help on programming, sorry).

 

[random260]

Yep fixed the the finance error on to year 2137.

Two workarounds
1. Ensure there are no available assistances coaches that can be recruited. So this bypasses ai signing coaches
2. (Most recent error bypass) copy and paste finance of older working copy to current copy

 

[bone2468]

Any updates comrades

 

[Lana]

Yes, but it's got a few steps. I won't go into it all now but here's an example:

Spoiler: Click here if you want an example calculation

A player's overall rating is the max of their rating in all positions: Inside Mid, Outside Mid, Ruck, Small Back Defending, Small Back Attacking, Small Forward, Tagger, Tall Back and Tall Forward. Each of these are different calculations.

For example, Inside Mid is weighted 15/5/20/35/25% to separate calculations for Tackling/FootSkills/GetToContest/HardBallGet/HandsInClose and then minus penalties for Endurance/Height/HardBallGet minimums (70/188/70). Tackling for example is weighted 35/35/30% to Aggression/ManMarking/Tenacity then plus bonus from improvement and minus penalties for degradation and particular injuries. Some of these calculations involve an age modifier which adds or subtracts from specific calculations.

So, to give a full rundown you can see I'll need to write a very big post!

Not sure why I didn't see this earlier, but thanks for explanation I can understand why you'd leave it at that. Is there a reason why players in AI teams seem to converge around the early 80s potential especially for midfielders?

 

[Simon_Nesbit]

Not sure why I didn't see this earlier, but thanks for explanation I can understand why you'd leave it at that. Is there a reason why players in AI teams seem to converge around the early 80s potential especially for midfielders?

The AI doesn't consider tomorrow. Every team is trying to win now. (This also leads to most clubs sacking their coach every other year).

Remember, we use "Positional Ratings" to value a player, instead of the actual raw skills of players. It's makes sense for easier comparison, but it's an artificial construct.

The game interprets some of a player's individual skills to provide their "Positional Ratings" (Tall Forward, Tall Defender, Inside Midfielder, etc). It's a mathematical calculation only and doesn't actually reflect all of the player's underlying skills or how those other skills impact on the match engine. Positional Ratings are based on specific raw skills (which don't change), plus "improvement" (experience) - which is generated based upon potential (Tall/Small), team selection (or maybe just games played) and age.

Computer team selections are then based on the players "Primary" position only (their highest rating), rather than best fit. My previous post referenced this issue - (picking a 65 rated "Small Defender" in the back pocket over a 75 "Inside Mid" who also has a PR of 73 as a Small Defender). So young players only get picked if they are in the best few in their position (no development selections). It's why mediocre/poor small defenders rack up 200+ games so regularly, and young players fail to flourish (no gametime). Computer teams will also pick stars at 60% fitness, so injuries create less opportunities than in human sides (well mine, anyway).

Similarly, the experience gained (Improvement) by players "appears" to be linked to team selection and position in the 22 (extrapolated from performance? or simply from matches played?) - however it works, it seems to create a positive feedback loop for the very best players (and any Tall forward); and a similar negative feedback for the Tall Defenders and small forwards. Aging is linear across all skills - easier to implement but perhaps not realistic.

So Computer opponents aren't picking the 18yo 73/95 outside midfielder they just drafted in the 1st round - at least not until they are the 7th highest rated midfielder in the team. If human coach picked that guy (a Top #10 pick), you are probably playing him off the bench immediately instead of the 74/74 32yo who's ready for retirement.

So computer players start later, and only get played in their primary position (as per the game's %) once they have developed naturally to be in the 22. That wastes much of their <21 years when the largest rating improvements occur.

There isn't much you can do to fix it unfortunately - I've tried changing club preferences and expectations to lower ladder positions, but that only seems to affect trading and coach retention, not selection. I also have a sneaking suspicion AI clubs either don't train or don't apply specialists - their gross improvement figures are much lower than mine.

 

[Lana]

The AI doesn't consider tomorrow. Every team is trying to win now. (This also leads to most clubs sacking their coach every other year).

Remember, we use "Positional Ratings" to value a player, instead of the actual raw skills of players. It's makes sense for easier comparison, but it's an artificial construct.

The game interprets some of a player's individual skills to provide their "Positional Ratings" (Tall Forward, Tall Defender, Inside Midfielder, etc). It's a mathematical calculation only and doesn't actually reflect all of the player's underlying skills or how those other skills impact on the match engine. Positional Ratings are based on specific raw skills (which don't change), plus "improvement" (experience) - which is generated based upon potential (Tall/Small), team selection (or maybe just games played) and age.

Computer team selections are then based on the players "Primary" position only (their highest rating), rather than best fit. My previous post referenced this issue - (picking a 65 rated "Small Defender" in the back pocket over a 75 "Inside Mid" who also has a PR of 73 as a Small Defender). So young players only get picked if they are in the best few in their position (no development selections). It's why mediocre/poor small defenders rack up 200+ games so regularly, and young players fail to flourish (no gametime). Computer teams will also pick stars at 60% fitness, so injuries create less opportunities than in human sides (well mine, anyway).

Similarly, the experience gained (Improvement) by players "appears" to be linked to team selection and position in the 22 (extrapolated from performance? or simply from matches played?) - however it works, it seems to create a positive feedback loop for the very best players (and any Tall forward); and a similar negative feedback for the Tall Defenders and small forwards. Aging is linear across all skills - easier to implement but perhaps not realistic.

So Computer opponents aren't picking the 18yo 73/95 outside midfielder they just drafted in the 1st round - at least not until they are the 7th highest rated midfielder in the team. If human coach picked that guy (a Top #10 pick), you are probably playing him off the bench immediately instead of the 74/74 32yo who's ready for retirement.

So computer players start later, and only get played in their primary position (as per the game's %) once they have developed naturally to be in the 22. That wastes much of their <21 years when the largest rating improvements occur.

There isn't much you can do to fix it unfortunately - I've tried changing club preferences and expectations to lower ladder positions, but that only seems to affect trading and coach retention, not selection. I also have a sneaking suspicion AI clubs either don't train or don't apply specialists - their gross improvement figures are much lower than mine.

Probably something that would have need to be done from the ground up. But, being able to customise the logic of how the assistant selects feature was one of these features that were underdeveloped. Being able to bias the rating due to age, select minimium fitness rating, lock particular players into particular places in the 22 (if fit) exclude particualr players from being selected at all etc. I never drafted in taggers for this reason. Small forwards too. Easier just to bring in inside outside mids with Tall forward. This feaute could have been used to help the non player teams have coherent selection policies.