Mooseton
Joined: Sep 6, 2010
• Posts: 593
Thanks for this post from:
December 1st, 2019 at 8:57:03 AM permalink
Perhaps 5dimes is following here. Lines are now adjusted to: Winner +475, No winner -650
\$1700, 18, 19, 1920, 40, 60,... :/ Thx 'Do it again'. I'll try
ksdjdj
Joined: Oct 20, 2013
• Posts: 841
December 1st, 2019 at 9:22:16 AM permalink
Quote: Wizard

Quote: ksdjdj

Quote: Wizard

... Please give my Lottery Jackpot Ticket Sales Calculator a try. I welcome all comments.

I just did a bit more testing and realized I get a different figure to you for your Powerball "probability of winners(s)" figure
I think your calculator is using the "Mega millions" chances for when you hit the "Power Ball" button
For example: when I type in "Jackpot size (in millions): 1000" and then click the "Power Ball" button, it says: "Probability of winner(s): 71.52%" ***
***: when i do it manually with the same tickets sold figure I get, ~72.76%. (as the "Probability of winner(s)" figure)

Note: For Mega Millions I get the same "Probability of winner(s):" figure as your calculator (after rounding to the nearest 0.01% )

For a jackpot of a billion, my calculator gets:

PowerBall: 71.52%
Mega Millions: 88%...

When doing manual calculations I still get the figures below:
1. With 379.99 million tickets sold I still get 72.75...% estimated chance for hitting the Powerball (about 1/292.2 million chance*** per ticket)
2. With 247.56 million tickets sold I still get 55.876...% estimated chance for hitting the Mega Millions (about 1/302.6 million chance*** per ticket)

***: I still think the calculator is using ~1/302.6 million chance for both calculations ?
Note: for "1" and "2" above, I used 1000 (a billion) as the Jackpot size,when using the "ticket sales calculator".
Also note: the calculator gets 55.88% for Mega Millions, so that is why I said before, that we get the same chance figure to the "nearest 0.01%", for that one.
Last edited by: ksdjdj on Dec 1, 2019
ksdjdj
Joined: Oct 20, 2013
• Posts: 841
December 1st, 2019 at 9:53:13 AM permalink
Quote: Mooseton

Perhaps 5dimes is following here. Lines are now adjusted to: Winner +475, No winner -650

Yes, anything is possible.
Personally I think it is normal, as 5d seem to "reset" the prices every so often ( I have seen it plenty of times before).
Note: I was actually very surprised that the "no prop" was +EV at such a "low" jackpot (in the past, a 250 to 350 million jackpot was the point to start looking for potential EV for the "no prop")

----
Bet Summary (update)

Game: Mega Millions
Book: www.5dimes.eu
Prop: "No Jackpot Winner 12/03/19"
Odds: -650
Estimated Tickets: 17.35 to 19.35 million (or \$34.7 to \$38.7 million worth of tickets)
Estimated Chance: 93.805...% (based on 19.35 million tickets)
Estimated EV: +8.23...% (based on 19.35 million tickets)
"Staking Method": 1/3 Kelly in total (I already had "half" my bet at -500, so I will use 0.5 x 1/3 Kelly for this part of my bet)
Bet Amount: ~ 17.8% of bank roll (or ~8.9%, if you already had "half" your bet at -500, like me).
Will I have a bet: yes

-----
Game: Powerball
Book: www.5dimes.eu
Prop: "No Jackpot Winner 12/04/19"
Odds: -1200
Estimated Tickets: up to 14.75 million (or \$29.5 million worth of tickets)
Estimated Chance: 95.07...% (based on 14.75 million tickets)
Estimated EV: +3.00...% (based on 14.75 million tickets)
"Staking Method": 1/3 Kelly
Bet Amount: ~ 12% of bank roll
Will I have a bet: yes (not trying to "outsmart" potential line changes anymore, so I am having all of my bet on now).
Wizard
Joined: Oct 14, 2009
• Posts: 21512
Thanks for this post from:
December 1st, 2019 at 4:53:00 PM permalink
Quote: ksdjdj

***: I still think the calculator is using ~1/302.6 million chance for both calculations ?

You're absolutely right. Thank you for the corrections. Perhaps I misunderstood your point before.

I just fixed it, please give it another try.
It's not whether you win or lose; it's whether or not you had a good bet.
Wizard
Joined: Oct 14, 2009
• Posts: 21512
December 1st, 2019 at 5:13:16 PM permalink
Quote: ksdjdj

Game: Powerball
Book: www.5dimes.eu
Prop: "No Jackpot Winner 12/04/19"
Odds: -1200
Estimated Tickets: up to 14.75 million (or \$29.5 million worth of tickets)
Estimated Chance: 95.07...% (based on 14.75 million tickets)
Estimated EV: +3.00...% (based on 14.75 million tickets)

Let's have a look:
Next drawing: Dec 4
Jackpot: 120M
5 Dimes lines:
Yes +775
No -1200
Probability of winner: 4.93%
Fair line on no: -1928

So, we're dead on. Did you do your own math on the probability of winning?

I think I'm going to bet \$100 per 1% of advantage. So \$300 on this one.
It's not whether you win or lose; it's whether or not you had a good bet.
ksdjdj
Joined: Oct 20, 2013
• Posts: 841
December 1st, 2019 at 6:23:49 PM permalink
Quote: Wizard

...I just fixed it, please give it another try.

Looks good to me : )
ksdjdj
Joined: Oct 20, 2013
• Posts: 841
December 1st, 2019 at 7:28:55 PM permalink
Quote: Wizard

... Did you do your own math on the probability of winning?

I used your calculator to work out an "upper ticket estimate" for the next PB draw (I think i also used it to work out the "middle" figure for the MM).
I rounded the figure in your calculator to the nearest 0.05 million

Reasons why I use or like using your calculator:
1. I like to do things the easiest way possible (probably like a lot of people),
2. I think it is accurate, especially for jackpots under \$450 million (I know that data is limited for bigger jackpots).
3. It seems to be very close to what I would get as my "middle or upper" limit in ticket estimate(s)
(I like my estimate to be on or a bit over the actual ticket sales when betting on the no jackpot winner).

Lastly, if i didn't use your calculator for this PB draw, I would have probably used 13 to 14.5 million tickets, so about a 95.15..% chance of no jackpot (using the 14.5 million ticket estimate)
Last edited by: ksdjdj on Dec 1, 2019
ksdjdj
Joined: Oct 20, 2013
• Posts: 841
December 2nd, 2019 at 2:02:32 PM permalink
The "no jackpot winner" for MM has gone from -650 to -550 since my last "bet summary" post about it.
The "no jackpot winner" for PB has gone from -1200 to -1165 since my last "bet summary" post about it.
Note: The "recreational punters" must be backing the "yes" side again.

----
Expanding on the "Kelly staking method" , mentioned in an earlier post, the steps to work out a "Kelly staking amount" yourself are below
Note : I am using MM draw on the 12/03/19, and 93.8% as the estimated chance of NOT being won
Note 2: : Kelly = EV / "odds" and then divide that amount by 3 (if you want to bet "1/3 Kelly" like me)
Note 3: "odds" are expressed as " to one" (so must be converted from American odds)

1. Work out/look up an "estimated chance" for the "No jackpot winner" (in this case ~93.8%)
2. Find out what the current odds are in "American" (currently -550)
3. Work out the estimated RTP*** by multiplying the figures from step "1" and "2" together (93.8% x 650/550)
***: RTP means estimated "return to player"
4. Take away 100% from the figure obtain in "3" above to work out the EV (110.85...% - 100% = +10.85...% EV)
5. Work out what the "full Kelly bet" should be by dividing the edge by the odds (10.85% EV / 0.1818... = 59.7%)
6. From the above, you then divide it by 3 ^^^ to get "1/3 kelly" (59.7% / 3 = 19.9%)
^^^: You don't have to bet "1/3 Kelly" I just picked that to make the bet "safer", you can bet 1/2 or 1/4 "Kelly" as just some other examples of making the bet "safer" in the "medium term" ( as long as you never bet MORE than "full Kelly")
7. The figure obtained in step "6." is what you should bet (as a % of your current "betting bank roll")

So from using the above steps, you should now be able to work out how much to bet in total

Important: In "step 3/4" I used a way I find easier to calculate the EV (the more "universal" way for working out EV would be below^*^):

("chance of winning x odds (to one)" ) - "chance of losing" = EV
(93.8% x 0.1818...) - 6.2% = 17.05454... - 6.2% = 10.85...%

^*^ You MUST use this method (or an expanded version*^* of it) for:
(i) Any game/sport where your bet has a chance of a "push/tie" (baccarat, is just one example)
(ii) Any game/sport where your bet can be increased or decreased ( "let it ride", is just one example)
(iii) Any game that pays "increased odds" the more "unlikely" a player hand is ("let it ride", is one example of this as well)

*^*: The "Banker Bet - 8 decks" table in the baccarat link is one example of point (i) and the "Let it Ride — Standard Pay Table" (just under "Analysis") in the let it ride link is an example of points (ii) and (iii) (see below)

https://wizardofodds.com/games/baccarat/basics/#toc-Odds
https://wizardofodds.com/games/let-it-ride/

----
Hope this is easy to understand and helpful.
Note: if it is not "easy to understand" , maybe someone with better communication/writing skills can write a better one in a new post.

(Even I think this is "Tl;dr", lol)
Last edited by: ksdjdj on Dec 2, 2019
Wizard
Joined: Oct 14, 2009
• Posts: 21512
December 2nd, 2019 at 4:52:07 PM permalink
Quote: ksdjdj

I used your calculator to work out an "upper ticket estimate" for the next PB draw (I think i also used it to work out the "middle" figure for the MM).
I rounded the figure in your calculator to the nearest 0.05 million

Reasons why I use or like using your calculator:
1. I like to do things the easiest way possible (probably like a lot of people),
2. I think it is accurate, especially for jackpots under \$450 million (I know that data is limited for bigger jackpots).
3. It seems to be very close to what I would get as my "middle or upper" limit in ticket estimate(s)
(I like my estimate to be on or a bit over the actual ticket sales when betting on the no jackpot winner).

Lastly, if i didn't use your calculator for this PB draw, I would have probably used 13 to 14.5 million tickets, so about a 95.15..% chance of no jackpot (using the 14.5 million ticket estimate)

Thank you for taking another look at it.

It's pretty obvious there is an exponential relationship between jackpot and ticket sales up to about 400M. Then it gets to be more of a judgement call how to draw a curve through the data. I can provide the exact formulas in my calculator if you're interested. For Power Ball there are actually three formulas, according to the jackpot amount.
It's not whether you win or lose; it's whether or not you had a good bet.
Wizard