how many of your days in % do you have a - all in EV day?

Wild guess: 50%!?

A winning player is more likely to run above EV over any given sufficiently small sample (such as a session) than below. When you get your money in and win you will always be running above EV unless you had 100% equity. So if you get your money in as a 55% favorite on average, then you will be running over EV over a single sample 55% of the time. For the same reason, losing players are more likely to run below EV over any sufficiently small sample. The magnitude of the EV evens out in the longrun, not the percentages of wins and thus times you are above/below EV. Eg if somebody somehow got their money in as a 99% favorite on average then out 99/100 samples they'd be running above EV. And if somebody got it in with 1% on average, they'd be running below EV 99/100 samples.

This part is true, but what you are missing is that you will be up or down <i>differing amounts</i> of EV - winning happens more often, but moves above EV less than losing moves below (if you change to 80% favourite you can visualise this pretty easily). Which means you will still be neither more likely to be above or below EV.

On rereading you weren't saying contrary to that anyway, right? It seemed misleading at first.


I assumed he asked a more meaningful question and wondered how often wiz's EV line was ahead of $0 when allin, not how often he won more than EV when allin.

I'd assume that the sample is v. small for this to be meaningful. I doubt that makes much of a difference after 2k hands shortstacking mostly being 60/40 or 80/20 equity. For someone like needbeer who is playing 15k hands a day, I would say there would be no significant statistical difference from 50%.

disclaimer: not a huge stats guy

It depends on how large your edge is and how often you are going all in.

Here's a simple applet that can do the math for you automatically using a binomial approximation: http://stattrek.com/Tables/Binomial.aspx

So for instance let's say you do 100 all ins with an average edge of 65% assuming all pots are for the same size. Your expectation is obviously 65 successes. Plug in the math on the applet and you'll see that you'll be running at greater than or equal to EV 55% of the time and below EV only 45% of the time.

Even with 1000 trials at 55% edge you're running greater than or equal to EV 51.3% of the time.

I rely heavily on my HUD when SS'ing, not nearly as much when FS'ing. In theory, yes, this would throw off the data an opponent has on you, however not many people play both 20-50bb tables as well as 40-100bb tables from my experience. If you happen to "full stack" the 20-50bb tables with a 50bb stack, your stats really won't vary all that much because you're generally surrounded by short stacks anyway, making your effective stack very close to 20bb in most hands anyway, and thus not effecting your stats all that much.


I'm a huge college football fan, I get turned off by a lot of the NFL drama so I don't follow it quite as closely. In other words I don't have any comment about Randy.

I do watch Indy play regularly though, because one of my good friends is starting safety, Melvin Bullitt. But since he was injured earlier this season, I've pretty much lost interest watching NFL at all.


I'd definitely do it, and at even money, but I would need significant action to make it worth my while. I'd want $50-$100k in action or so (not sure who we could trust to escrow that much). I personally don't think I can get that much action, even with 1:1 odds, but if you're more than welcome to raise interest for me if you'd like. =)

There are a few problems however, the 40-100bb $5/10 doesn't run a great deal. At night time there are only a couple tables running, and at peak hours there are still usually less than 10 tables running. So, if I were to average like 4-6 tabling, it would take a long time to get a couple hundred thousand hand sample. This sucks for me, and the reason I'd require so much action, because I could just SS for $50k+ a month and make better use of my time. Another problem is many people have issues with tying up their money in a prop bet for lengthy period of time, especially when it's a large sum.

I don't feel I have anything to prove to FS'ers or SS'ers. It would be pursued solely for the money because it's +EV imo. If you think there is significant interest, then I'd love to do it.


1. When I filter by stack size I become more and more profitable as my stack size increases. This has a lot to do with me sticking around when there are fish present at the tables.

2. I don't necessarily think it's better. Sometimes I like to have them covered, it depends on the type of fish imo, there are so many different types. I like having the callstation fish covered. But I like SS'ing against the uber agro preflop spew fish.

3. I'm sorry, I've decided not to go into any specific SS'ing stat based strategy discussion in my well, my students pay me well for this type of information. I hope you're not offended in the least.


Best: +$55k (happened twice)
Worst: -$75k


I'm glad to hear it, someone else recently mentioned my old database as well, and I was very happy to see he was still around. I hope you have continued success sir.

I made a pretty decent name for myself fullstacking (yes, fullstacking) the highest stakes on PartyPoker before they closed their doors to U.S. players. I made a very substantial amount of money in 2006 and decided to sell the last few months of my database here on 2p2. I sold several hundred copies. Deja vu.. I remember several regs being upset that I was sharing this information.. looks like some things never change.


What Aillemkall said, I generally play more hands in the blinds and less hands from the other positions, pulling the average down closer to 2bb/100 or 1ptbb/100.


Biggest downswing was $220k in a little less than a week (but this was mostly $25/50 and $50/100 HU, so it's not a completely unusual downswing). But my confidence was a little shaken, so I took a couple weeks off, spent some times reviewing hands, planning, etc. Then I came back fresh, moved down in stakes, and quickly rebuilt my confidence.


My specs aren't absolutely blazing compared to the latest technology but it runs everything just fine. I don't play on my laptop.

OS: Windows 7
Processor: Intel Quad Core 2.66GHz
RAM: 8 GB
Graphics cards: 2 BFG GeForce 7800 GTX OC
And obviously a motherboard that supports two graphics cards.
(most importantly you'll need a couple graphics cards with a pair of Dual Link DVI ports)


There is no way I can accurately answer this. Closest I can come is to say.. what is his expected income playing poker using optimal table selection and only when he's on his A game... is that $60K a year? If so, then if he can make more than $60K a year pursuing SNE after his losses at the table, then I suggest doing that. Even if it's a little under $60k, I would suggest SNE even though it's more time consuming/stressful, because it will pay future dividends with the SNE 5xVPP multiplier the following year.. not to mention the enjoyment of travelling with the SNE packages /vacations and the awesome suites they put us in.


I made $55k in rakeback last month, so I have a little trouble relating to the question. But I suppose that I would play poker until my income barely covered my bills before I'd consider working for a corporation again. But likely before that happened I would explore other means of income (starting my own business.)


Script joins wait list and if a table becomes available that I can't join with 20bb, then it can close that table out. You can get this from my programmer Brian. He can create any script you can dream up, he's legendary.

He’s asked that all inquiries be directed to his 2p2 account, brianwi. He asks that you PM him your AIM screenname.


I'll be honest, I stopped paying attention to EV earlier in the year because I was was running quite poorly and I found it as a source of frustration. I'm fairly confident in my abilities as a poker player and don't feel the need to check EV for reassurance that I'm playing well anymore. I've been much happier since I eliminated it from my HEM reports. I understand that this isn't easy to do or even an option for a player that is struggling and monitoring their results closely, however, if you are checking regularly, don't make the mistake of checking too frequently.. EV should only be checked once a week imo, even if you're grinding SNE type volumes.

How do you feel the legalization/regulation/taxation of online poker would affect the profitability of online poker?

Thank you very much for doing this. You story rly motivated me to work hard and try to be that successfull myself.

- Why do you play FR now? (just for lower variance?)
- How do your charts work? (not asking for specific infos, just the general idea behing them)
- Had you something like a mentor when you started or someone who inspired/still inspiring you?
- Are there any goals left poker related you that you want to archive?
- Any hints to maintain a succefull relationsship when grinding that much?

When playing out of the blinds, how do u adjust to fish that call too much and call of with like 2-3rd pair too much? How about players who start to float your cbets?

Are you aware of your standard deviation being 44bb/100 ? Did you ever had >20k hands sample where your std dev bb/100 was lower than this?

Are you then aware how differently you were playing generally purely looking at the std. dev bb/100? FYI mine is 26ish

Hi, I hope you will understand my questions, because my English is not as well as I want

1. You wrote here (if I understand it correctly) that a good regular player is a player who knows (exactly) how to play against a concrete player, I mean having a read on him. Are you able to use this reads when you´re playing 24-tabling? You said you have only about 1 second to act (approximately), so it´s able to use them? Or, which kind of notes or read are you used to? I know that PT can get you an information about the aggresion and so on, but you have to got reads after playing with many player so often.

2. Do you think that the poker software like Poker Tracker is absolutely necessary also for the intermediate players (not a beginners, not a pros) as I am, who playing, for example NL20? Do you think that a player can be successful without using any kind of the programme like this or it is necessary? I mean, it cost, i think 80 dollars, and Im asking to myself if its a good invesment for me.

3. Are you using dual monitors and having a one single computer or are you using dual monitors with two computers? I really dont understand how can you use dual monitors on one PC. Can you specify this?

4. How can you recognize ´fish´ player? What they doing wrong? Are they often shortstacking or what? Are they having a different stats than a ´normal´ regulars?

At last, Id like to tell you that you change my point of view on playing poker, seriously. This whole thread is like a teaching book to me, it read there much more important (and interesting) informations than in many ´quasi-strategy´ articles on the web.

I wish you a good luck and enough power not only to beating the players, but also to beat your disease!

Lusodas (from Slovakia)

#264

Edit on that. my std dev bb/100 is 39bb/100 but still I am pretty sure your game is significantly looser, more hi variance just for the sake of you grinding out millions of hands and I am quite curious if the std. dev. per ''period'' (month, 2 months or per 3 months) stayed the same in your poker career

Assuming your health improves when do you think you'd like to get out of poker?

Do you feel it reduces your variance if you leave at 30bbs as opposed to staying and building ur stack?

Also, do you feel that it makes it more easy and efficient to grind out hands leaving at 30bbs as opposed to staying and building ur stack?

Thanks for the well!

This just doesn't make too much sense to me. What is it about checking EV and winnings that frustrates you that doesn't frustrate when you just check only winnings? I know you've said you check winnings more than you'd like to, so maybe it was just easier to stop looking at EV than it would have been to stop looking at results altogether?

Personally, if I were to restrict myself from looking at either winnings or EV adjusted winnings on my tracker, I would choose winnings. After all, EV adjusted winnings is just a less noisy and less luck-influenced version of your actual winnings. I guess my point is, if you're not going to look at EV, why look at results at all?

i have to agree- that's what i did recently...just put all ev stuff in my HEM and took out all winnings and non-ev bb stuff

Non all in preflop EV results as a result of actual expectation are going to be very misleading for shortstackers.

Without getting into a lengthy reasoning why just apply reductio ad absurdum. Let's say a 100bb player opens AA to 99bb. He gets called by another 100bb player who has 33. The flop is 953. They get it all in for one more bb on the flop. The turn is an A, river is a blank. EV is going to show that play as being very +EV for 33. Which it obviously is not. If you repeat that play 100 times 33 is going to be down about 80 buyins, even though if you rely on EV it would say he would be up 90 buyins.

The larger a percentage of stacks that goes in preflop, the more misleading EV results are. For 20bb ratholers, even 2bb is 10% of stacks so this bias is huge. It tells you at a snapshot of that hand that you were +EV while ignoring all other actions. So looking at an EV graph and it saying you 'should be running at x bb/100 in the long run' is not the correct interpretation. It fails to compensate for the EV of plays you make at the points leading up to where it calculates the EV so very -EV play can and often is considered +EV, and vice versa. It basically picks a completely arbitrary point to calculate the EV of the entire hand. Whether you're running good or bad from that point says nothing about the EV of how you played the hand as a whole.

And it doesn't end at non-preflop all ins. Now for card removal! Say UTG opens, gets flatted by MP and CO. A ratholer in the sb shoves AQo. UTG shoves/calls with JJ, everybody else folds. Ok, AQ is a 43% dog so +EV. But the reality of card removal is given that action he's actually often going to be a 38.5% or even worse dog. And this doesn't even out in the long run. Consider your range of highcard hands for shoving and what you're getting called by, vs your range of non highcard thats shoving and getting called by highcard/broadway type stuff. Once again, the EV result will be substantially far off of reality. Obviously I am not suggesting shoving is not +EV here, but simply that the 'EV calculation' for when you get called is often going to somewhat dramatically overstate your amount of equity.

I think there is good reason to believe that straight up winnings are actually likely to be more accurate than EV for ratholers. EV introduces too many biases and is very inaccurate for what it tries to measure. There's a reason the vast majority of ratholers run below EV, and as much as I'd like to say it is - its not because of karma.

I ran at break even/ slightly +EV (ie getting luckier than I should have) for my 1.8 million hands. Since then horrible EV run for the last two months and am slightly under (making a whopping 2-3 dollars less per day than I "should have" over the last 3-4 years), but it seems pretty close to accurate in my case at least.

Nah, false assumption when pertaining to me. For one, I filtered out 20bb exactly because your winrate will clearly be lower at this exact stack size. I also review my results by position only. Mama didn't raise no dummy, curt.

I can see potential pros and potential cons, all depending on how heavily it is regulated/taxed. I think legalization would be great for giving the game more exposure to amateur players, however this benefit will be offset if taxes are increased greatly and we have to eat these taxes through a significant rake hike.


1. I'm not playing much full ring at all actually, I usually play shorthanded until the table fills all the way up. Xpert 6xVPP move imo. =)

2. My HUD is color coded to my charts. The various stats/colors reference particular areas of the charts. I find this method the best for memory and quick recall when determining the appropriate hand ranges to play in certain circumstances.

3. No, I started playing and motivating myself a couple years before poker was so popular on the internet (the Moneymaker "poker boom".) There wasn't many (if any) online resources, coaching, or really any online mentors at the time. So I've pretty much become accustom to teaching and learning by myself over the years. I will say that in 2009 when I took interesting in SNE, and I found out that George Lind had earned a staggering 3 million VPP in 2008 that I was very impressed. I amazed that someone could earn so many VPP.. and seeing all those milestones added up was exciting. So I decided to give it my best shot! So I think you could say George was inspiring, even though I wasn't a SNG player, and knew little about him, just to see that he had such dedication was an inspiration to me. Records will continue to be broken, because we continue to show others what is achievable through hard work and dedication.. and people come along every year that push themselves further than they even thought possible, myself included. I'd like to think I've inspired several of you out there to surpass my record. The record isn't something I covet as comparison of my abilities to the rest of the population, I know many of you are capable of surpassing me if you fully dedicate yourself to the pursuit. I pursed the record because I had something to prove to myself, and now that I've done that I am very proud that I passed this test of attrition. And I know whoever takes the record next, will get to enjoy a similar experience, as anyone who has worked this hard should.

4. As far as online is concerned, I haven't put much thought into a new goal as of yet, but I most certainly will, as they are my primary driving force. And as far as live poker is concerned, I'd like to get healthy enough to travel again and final tabling a major event some day would be an incredible experience.

5. Don't bring work home with you. In other words, similar to coming home from a stressful job you need to leave any potential frustrations and negative thoughts at the door. Before you walk out of your office, give yourself a moment to unwind.. sometimes I even find this necessary after a good session, because I still might be in poker attack mode. If you find it necessary to vent your frustrations, vent to your coach, your poker buddies (though they may to tell you to shut the hell up lol), or go post in BBV or something... just don't bring it up to the woman in your life. Your relationship will be much more successful imo if she isn't aware of all the swingy ups and downs that we have to endure. As long as you aren't squandering the entire family income and have a sufficient bankroll to withstand variation, then you should just man up and squash your poker frustrations at the door. Whenever you get bent out of shape, just remember that you have the life of a poker professional while much of the world is starving or having their face blown off by an artillery shell. Just enjoy your time with her because those moments are fleeting.



I'm sorry, as I mentioned before I'm not discussing questions related to strategy.. my students pay well for this information and it wouldn't be fair to give it away freely.


1. Well, in my opinion it's much easier to get reads the higher up you move in stakes. Much of the decisions I make are statistically based, and when dealing with a much smaller player pool this becomes an advantage to my playing style. The less unknown players I have to deal with, generally the better, unless the unknown turns out to be a huge fish. Also, I am a big supporter of taking color coded notes in both Stars and Hold'em Manager. These notes allow me to make quick reads on an opponent's tendencies based on the color combination that I have labeled for them. And again, the smaller the player pool equals the more notes that I have, and the more accurate my reads are. Many times that I sit down to play a high stakes session, 4 or so of the 6 guys at the 6max are almost the same people at every table, so it doesn't really feel like I'm playing 24 tables at all. It just feels like I'm playing a few tables worth of players very quickly. Don't get me wrong, I like facing unknown fish and have a particular ABC poker technique I approach them with, it's just that the majority of them don't require any significant reads. So if you are playing at a low level with many fishy unknown players, you need to focus more on playing ABC poker (getting good value out of your good hands, not leaking money with hands that show down poorly, not chasing weak draws, etc), then necessarily focusing on particular reads for these opponents. Many of them play so irrational without any true rhyme or reason for their actions that trying to make a read is near impossible anyway.. so your read should be more focused on your hand and its showdown value relative to the board.

2. I always think Hold'em Manager or PokerTracker is a good investment for understanding and evaluating your play. It's also great for studying the play of another regular. As far as needing a HUD to play a game like NL20, no I don't feel it is necessary if you're playing just a few tables because it's easy to follow the action, however, if you're going to be mass tabling then I think a HUD is very valuable no matter what level you are competing at. It makes your ability to make quick decisions much more efficient and effective.

3. I have multiple monitors with one PC. A decent video card will have two outputs for monitors, so it only takes one of these if you want 2 monitors. I recently upgraded to 5 monitors running through 3 separate video cards on one PC (which I don't have a screenshot of yet), because I've always felt the more monitor real estate the merrier since I have been known to tile up to 40 tables at once on occasion. It's my best option because I do not enjoy stacking or cascading of any kind.

4. Look at their VPIP stat and if it's very high then you know they are a fish. Also, yes, plain and simple if you see them do something unwise, you know they aren't a good player. But if they are short stacking, that means they are likely the best player at the table! ;-) Just kidding.. you can't make an accurate assumption on a players ability based solely on the amount that they buy in for.

Thank you sir, I'm glad I could help you and change your life in some way for the better. =)


My career has been a long one, and I've gone through dozens of databases so it's difficult to say what my std deviation has been over my lifetime and how it has varied. I'm sure it varies quite often, even within the same game (6max) over the course of a few months. The information I provided pertained to my last 6 months of 6max play. I short stack the highest stakes on Stars, so I expect my variance to be quite high. I honestly pay zero attention to my std deviation. I only looked it up because I was asked. I just play the best I possibly can every time I sit down at the tables, and I keep plenty of money in my bankroll, so I don't find any real reason to waste energy digging deep to investigate any variations in my std deviation.


I plan to have poker present in my life in some capacity for the remainder of my life.


Well generally when I stay beyond 30bb I am staying because I enjoy the dynamic of that particular table, be it a big fish / weak regulars, and that in itself generally lowers variance. So, I wouldn't say having a large stack is more variance for me... but in a vaccum if you gave me 20bb or 30bb or 40bb against the same set of good regulars, I think variance would be higher as stack sizes increased in a linear fashion.

I definitely feel it's easier to grind out hands/VPP's when leaving at 30bb. But better for my winrate if I stay around longer (as long as I'm playing my A game, if not it can be worse for my winrate to stick around beyond 30bb.)


I look at actual results because like the rest of you I'm curious if I won or lost money. Sometimes it's hard to fight that curiosity when I know I probably shouldn't look, and sometimes I have no real reason to fight the curiosity and I enjoy looking. EV is over a decent sample is good for evaluating your play, but let's face it, actual over a large sample is good for evaluating results as well. And since you can't cash EV out and eventually you're going to have to look at your actual bankroll anyway, then ultimately you're going to know if you've been running well or not.

Looking at only one would be fine for the most part though, no matter if it's EV or actual, I do see where you're coming from. But it's not really a point of debate for me. I was just saying that I personally find it a source of frustration when I compare the two and I continue to dwell on the fact day in and day out that I'm running poorly yet again or something of that nature. It's all on personal preference, this may not bother some people. I was just explaining how I pursue the situation.

I don't always check results, but when I do, I prefer Dos Equis XX.

This doesn't introduce any bias to all-in EV as calculated by HM. All-in EV calculates your EV when the money goes in, it doesn't claim to tell you anything about the way you played the hand or your overall luck

This is true. There is bias the other way when two deep stacks call a SS shove pre then one of the deep players bets post with their made hand (bias because the deep players are more likely to bet when they hit their hand.)

I doubt these are a large effect however, and becoming less so as the tables have less deep stacks.

Straight up winnings may become more accurate in the very long term depending on the size of those biases (I would suggest small.) However, you just have to run 50+ short BIs below EV a few days to see how far off actual winnings are in the shorter term (twice this week ffs)

It tries to measure EV at the point money goes all in, and I would imagine it's pretty accurate at that.

Do they though? Me and my mate are both above, but we probably don't find that as news worthy as someone 1000 buy-ins below does I'd be interested in seeing the true figures to see if there are any large biases.

Do it right:
The first part of your post is nonsense but you are right at the end about it not being perfectly accurate.

With the 33 example you are trying to get an allin ev calculator to do something more than that. It will be 100% accurate in that instance for what it does. It's like you are calling a machine that can only add flawed because it can't divide. Do you see hOow stupid your argument is? No one is saying it can do anything but that one simple task. Everytime you have 33 on the flop in that spot your allin ev calc is more accurate than not having it. To reiterate, the instance you mention details about that hand before being allin - you are talking about something this program doesn't claim to do. It's really surprising to always see these arguments when people are claiming something is flawed for not doing something it doesn't claim to do. Think about that for a second.

And if you continued with your argument of the program not being able to look at things from a step back where do you draw the line? Why look at that 33 hand when you even have just 33 why not your average hand? This can be taken back all the way which is to the win rate per hand per position. Obviously not very useful but "perfect". The only perfect ev calculators are that or an allin ev calculator. Nothing in between has perfect information.


Now when you talk about the ev itself being flawed you are in fact correct. Consider the most common shortstack situation; it folds around to btn who raises and the bb shoves and the vtn calls. Since 4 players folded worse than average hands the deck is not random. With more high cards left in the deck this favors the player with the better hand. So now the only question left to answer is who gets it in worse in these spots and therefore always runs worse than ev, shortstackers or full?

In the above situation in theory in players strategy should
Be the same. But if we add anither common spot, say the sb 3bet shoving instead we know that this will happen more often with ssers because they have less to risk with the bb left to act.

In summary, the first part of your post is moronic, the second genius.

Sorry to hijack needbeer

I never said the first point negates EV but rather that EV measures something other than what most people think it does. Look at the people above this thinking EV is more useful than $ since it takes out 'luck'. EV simply completely ignores all luck prior to the money going in and instead focuses all on all luck of the cards dealt after the money goes in. Far from taking luck out of the equation it magnifies one aspect of luck and makes another huge issue of luck completely invisible. $ earned balances all luck.

It'd be like if you added another line to your graphs that showed your EV considering only the river card even if you got it in preflop. So you'd have a new line and sometimes it'd be running good, sometimes bad it'd certainly be vastly less accurate than $ earned regarding your 'true' expectation. I think this should be fairly obvious yet for most people this is not obvious with regards to the current EV calculations.

And the second part with card removal is that not only does EV measure something other than what most people think it does, it does a poor job of it. EV is great to make you feel better about yourself after having a bad session or whatever but in the long run its a completely meaningless figure that does nothing more than do a flawed job of measuring some really arbitrary value.

Sorry, but all-in EV is most definitely a better measure than actual winnings in any sample less than say 100 million hands at least. The standard deviation on my all-in EV is about 25bb/100, whereas the standard deviation of on my actual earnings is around 42bb/100. This makes the standard deviation of the EV-difference approximately 34bb/100.

The first part of Do it Right's post:
EV does not negate all luck, hence why it still has a standard deviation. It does however negate more than 50% of the luck involved in the game (short stacking). I believe it would negate a much higher % of the luck full stacking but I don't have data to back that claim.

Second part:
There is a bias introduced here, but it's very small. For the most part, the all-in preflop range vs. range equities are very close to 50-50. In some spots it might get to 46-54, but you're not going to find a spread larger than that in most cases. Introducing a dead card bias will affect this extremely marginally in the long run if a player consistently puts it in bad or vice versa, but keep in mind it affects both players.

Estimating the bias:
For a 44.06 - 55.94 flip (21.6% shove range vs 12.2% call range), 7 random dead low cards (3s 4h 5d 6c 7d 2h 8h), change the preflop equities to 44.045 - 55.955. This is a bias of 0.015% in a fairly generous example. If the average total pot is ~45bb, this would introduce a bias of 0.00675 bb / pot, or 0.675 bb / 100 aipf hands.
Aipf is about 3% of all hands for me, so that this would affect the overall winrate by 0.02025 bb/100 in the long run.

Now the question is, over what size sample does the second measure become a more accurate predictor of the the truth:

1) A mean deviation of +0.02025 with a standard deviation of 25
2) A mean deviation of 0 with a standard deviation of 44

For the ev-difference, which has a std dev of 34 to drop it's standard deviation /100 to less than 0.02, you need a sample size of 289,000,000 hands (calculation: 34/0.02 = 1700, 1700^2 * 100 = 289 mil).

Wee I overshot my initial sample estimate by 3 times =)