For those looking for strategy tips on Vegas mode, definitely check out Pocket Gamer's in-depth guide: http://www.pocketgamer.co.uk/r/iPhone/Sage+Solitaire/feature.asp?c=67918 Helpful business.
Incredibly useful indeed. Thanks for the link; been trying those suggestions for a few days and starting to make progress.
Is there a guide for the standard Single Deck mode anywhere? I truly suck having won only 1 game out of 48 so far. Thanks in advance!
Try and trash a card from one of the big piles every second turn. You should be able to clear the board often, but will not get very high scores
True Grit Math Hey, as a degenerate gambler I've been really interested in True Grit mode. I thought I'd take a pass at answering some of the central questions I had about the math behind it. My main objective was to figure out the EV of the various bet-sizes. ("EV" is a standard gambling term which means how much money a bet is worth on average, in the long run. A +EV bet is a profitable wager, and a -EV is a losing wager.) After a few hundred hands I was fairly confident that the $50 and $100 bets were +EV for me, but what about the bigger bets? Were they more profitable? I had a vague suspicion that they were not, that the bigger board clear bonus would be offset by the fact that the size of the bet increased while the size of the average bank remained fixed. I thought it was likely these bigger bets were -EV, and that the people at the top of the True Grit highest bank score board were people who had rushed to these highest stakes, got lucky a few times in a row, but wouldn't be able to sustain this win rate over time. As you will see, I was wrong. The bigger bets are more profitable, in most cases considerably more profitable. Here's the basics of my method: I start with a couple of base rates - estimates of how often I clear the board and the size of my average in-game bank at the end of a game. These base rates don't change over the different bet sizes, and together they form the foundation of a player's overall "win rate". I'm estimating my board clear rate at 60%, and my average bank at $22. This board clear rate is much higher than I would have originally guessed, but I ended up here after looking at how much money I had made over ~300 hands and assuming an average bank of somewhere around $20. I could be off here. (Zach if you're listening I'd love to have these two stats in future builds, they're much more useful than most of the things you're tracking!) The other estimate built into these calculations is what kind of bonus I earn when I *don't* clear the board. Here I assumed clearing two of the top three rows and two of the middle three rows. Seems about right. Given these base rates, it's simple math to look at the EV of the various bet sizes. The Average Clear Payoff is the average bank multiplied by the board clear bonus for that level. The Average No-Clear Payoff is the average bank times the bonus you would get from 2 top row clears and 2 middle row clears. The Total Average Payoff is the Average Clear Payoff times the Clear Percentage and the Average No-Clear Payoff times the inverse of the Clear Percentage. (ie, if you clear the board 60% of the time, then your Total Average Payoff is 60% times what you make, on average, when you clear the board and 40% times what you make, on average, when you don't.) Here are the results I got: Couple of interesting things to note. The conventional wisdom that $200 was a sucker bet is confirmed here. It's better than $100, but not by much, whereas the $500 level represents a massive jump in profitability over the previous levels. Another surprise for me was how sensitive the results were to your average bank amount. I assumed that one should sacrifice almost any amount of bank to increase board clear rate, but in fact boosting your average bank by $1 (and keeping your clear rate the same) is far more valuable than boosting your clear rate by 1% (and keeping your avg bank the same.) Higher Clear Rate: Higher Avg Bank: It's interesting to see that the cutoff for profitability seems to be around a 50% clear rate and a $20 bank: So as long as you are clearing the board more than half the time you are probably a long-term winner. With these estimates established, the next step is to answer the question of when you should go up in stakes. Answering this question requires thinking about edge, variance, and risk of ruin; and will probably involve the Kelly Criterion. I will save that for another post. Let me know if you spot any mistakes in the above or have any questions. Also happy to discuss general strategy for increasing board clear rate and average bank, which are obviously the most important parts of the whole equation. /mc
If you're at that 60% win rate, doesn't it make the most sense to advance to the 500 and 1000 bets once you're able to do them twice in a row? So, start betting 500 at a $1500+ bank and betting 1000 at a $2000+ bank? I mean, if you're good enough to clear the board with a positive $ return more than half the time, you just want to be at a place where you can safely make those good returns bets twice in a row. That's the logic I've been using, at least. Has tended to work out. Definitely breaks a bit in the new bets Zach's releasing soon and won't follow the same algorithm for 10k+ bets
Short answer: no. Long answer: https://en.wikipedia.org/wiki/Kelly_criterion The Kelly Criterion is proven to be the fastest possible way to grow your bankroll. My guess is that you want to have around 8K - 10k before moving up to $500 bets. At some point soon I will try to figure out the actual thresholds. It's complicated and I'm bad at math. Games are fun! /mc
my mind is blown. BLOWN! i'm reading it. the math seems logical, but i refuse to obey. sticking to my gut here. edit: from the wikipedia article: so, not 8k-10k, as you suggest. if the true win rate is 60% and you're doing kelly bets you'd want to start making 500 bets at a bank of $2,500 and 1k bets at a bank of $5,000. right? or is the weird $$ return on your bets in sage causing this to not be so simple?
Yeah, I think the fact that there isn't a well-defined edge and odds here makes it more complicated. Here's one way to think about it: if we were getting 1 to 1 odds with a 60% chance of winning, then a $1000 bet would have an EV of $600. But according to my estimates, the $1000 bet has an EV closer to $240. 240 is 40% of 600, so let's use that as a rough guide. 40% of 20% is 8%. How much money would we need in our bankroll for a $1000 to be 8% of it? That's $12,500. This is the not the proper way to do this, but I bet it's not too far off. If it turns out to be correct, and generalizable across all the bets, then ~13x the bet size is the threshold for moving up. /mc
got it Well, that is freaking nuts. I tend to bet max when I can afford to lose it twice in a row. Was thinking about moving that to four times in a row for the extra punishing higher bets. This has me rethinking my entire betting strategy. I don't know if I'll fully commit to betting 1/13th of my bank since that seems too slow, but I respect it as a play choice and will probably be betting closer to that way now.
1.1.1 is out! New bets for pro vegas players (10k, 25k, 50k, and 100k)! Those bets also include a new mechanic, the lowest piles, instead of being worth 0, are worth negative multiplier. So be careful about clearing them if you don't think you can clear the board. also some bug fixes.
This sounds really fun but I'm nowhere close to that yet Every new update keeps me playing more so I'd love to see them keep on coming!
Score!! I hope the update also eliminates the way you rig the game. Because the game is rigged and you = rigged it. Seriously though, it baffles me why someone would think that. As you point out in the update notes, it would make game design less interesting and there are no financial incentives in this case. Keep up the great work! And remember to update your App Store reviews, TAers.
No financial incentives?! I refuse to buy a premium upgrade for a game that doesn't even want me to play it! I've been watching the fours and they are all rigged. I don't know how the dev though they could slip the fact that a top deck four is always followed by a sequence of 7,8,6,9,A,4,4,J,7,2,7,5,J,4,7,4,7,4,7,J,A,2,Q by us!
Anecdotally, I've gone on 12-bet size downswings before. That lends some intuitive evidence for wanting a bigger ratio of bankroll to bet size. /mc
Any hope in a future update that makes Vegas Mode a bit easier or more forgiving? I've been working on it for weeks and haven't gotten higher than $650. I read the pocket gamer article but still struggling. Really really frustrating.
* Big 2.0 Update Just Released * Hey everyone! I just wanted to drop in and give you a heads up that I just released a huge update to Sage Solitaire! The big thing we added was Insta-Tournaments (From Pocket-Run). Basically you get to play a Single Card Deck against the world and see how you stack up against everyone else who played that exact deck! But it's actually a little more than that — If you've been playing my games for a while you know that every mode a significant twist on the core game, Insta-Tournament in Sage keeps this trend going with a devious new scoring mechanic called 5 Card Bonus! Every time you make a hand, your bonus goes up, and every time you trash a card it resets. The bonus only gets applied when you make a 5 Card Hand (or a 4 of a kind), so you'll want to keep the hands constant and the trashes to a minimum. It adds a good amount of risk/reward to the game and pushes you towards doing sub-optimal things (like mixing flushes and straights in the same game). I've been surprised at what a nice little twist it is, and I hope you enjoy it as much as I have. Also I hope you play a bunch so I have more opponents in Insta-Tournament
Hey Zach, thanks for the new update! I'll definitely have to try the Insta soon. Just wanted to let you know that, at least for me, changes made to wallpaper design reset every time app is reopened—mine keeps reverting to an orangish wavy wallpaper. Obviously not a game breaker but thought you should know. I'm representing with the iTelephone 6+ running the latest/last iOS 10. Thanks again and rock on.
