20-25% chance each month - does this compound??

PrettyPalomino · January 2016

I'm having a hard time wrapping my mind around this probability question, it seems simple up front but I've dissected it and now I need help getting out of this black hole.

Let's say you have a 20% chance (1 in 5) each month of conceiving. You don't conceive the first month. Does that mean you now have a 40% chance of conceiving in month #2?

If 1 in 5 people TTC successfully conceive each month, shouldn't the odds of success by month 5 be close to 100%? (all health/fertility factors aside)

I realize this is borderline BSC and maybe I do want someone to say "YES you are in month 3 therefore your odds this time are 60%!"

Any stats wizards out there???

MrsDho11 · January 2016

Hey @PrettyPalomino, I'm feeling lazy. Here's this:
https://en.wikipedia.org/wiki/Gambler's_fallacy

Of course, this applying assumes that the odds of each month of conceiving are independent...

Edit for grammar. Maybe. I'm tired.

simcal18 · January 2016

No. You have a 20% chance each month independent of how long you have been trying (unless you have fertility issues). So by month 2, 64% of healthy couples still won't be pregnant. By month three that's around 49%. By month 6 around 74% of healthy couples should be pregnant, but that still leaves 26% that won't be.

LoveIsOwlYouNeed · January 2016

PrettyPalomino said:

I'm having a hard time wrapping my mind around this probability question, it seems simple up front but I've dissected it and now I need help getting out of this black hole.
Let's say you have a 20% chance (1 in 5) each month of conceiving. You don't conceive the first month. Does that mean you now have a 40% chance of conceiving in month #2?
If 1 in 5 people TTC successfully conceive each month, shouldn't the odds of success by month 5 be close to 100%? (all health/fertility factors aside)
I realize this is borderline BSC and maybe I do want someone to say "YES you are in month 3 therefore your odds this time are 60%!"
Any stats wizards out there???

I think it just means that any month you have a 20 percent chance. It doesn't double the next month and doesn't mean that in five months you're at 100 percent. It basically "resets" to 20 percent every month, if it helps to look at it that way.

ETA: Also consider, if you were 100 percent by month 5, you'd be guaranteed to get pregnant the fifth month and thus the whole up-to-a-year thing would be null and void.

bsckgb7 · January 2016

No. Doesn't work that way. If you are pregnant, you have a 50% chance of having a boy. The next time you get pregnant doesn't mean you have a 100% of boy if you didn't have a boy the first time.

Each month is a separate entity. They aren't dependent on what happened last month.

nlwz123 · January 2016

No. You have a 20% chance each separate cycle.

MissAmeliaPond · January 2016

I think PPs have covered it but I'll just add another example. It's like playing the lottery: your odds of winning do not increase each time you play, they're the same every time you buy a ticket.

You want it to be the "Monty Hall" problem scenario (I did too), but it's just not. Each month is a discrete chance.

MrsDho11 · January 2016

I will not nitpick. I will not nitpick.

mamadomino · January 2016

Does anyone know what goes in to the "20% chance with perfect timing" thing? Is that once in the fertile window? Right before o? I always have so many questions about what that means.

DumbledoresArmy · January 2016

It's funny because I was just thinking about that last night. Logically I know that it resets every month, but I can't help but think after every cycle "odds are I should get pregnant this time!!"

MissAmeliaPond · January 2016

@MrsDho11 you can nitpick if you want.

simcal18 · January 2016

MrsDho11 said:

I will not nitpick. I will not nitpick.

Nitpick away @MrsDho11! Would love to hear from someone who has more experience in statistics than I do. The way I approached it is that if a couple has a 20% chance of becoming pregnant each cycle, there is an 80% chance that they will not become pregnant. So to calculate the probability that the couple still won't be pregnant after 2 months (assuming good fertility and assuming all other variables stay steady, which is rarely true): .8 * .8 = 64% probability of no pregnancy after 2 months. For three months, .8^3 is about 49%, for month 6 .8^6 is around 26%, and so on and so forth. But maybe I'm missing something.

MrsDho11 · January 2016

mamadomino said:

Does anyone know what goes in to the "20% chance with perfect timing" thing? Is that once in the fertile window? Right before o? I always have so many questions about what that means.

I can't put my finger on the exact origin of that right now, but I believe that it is roughly "if you had sex every 2 to 3 days during your FW".

But that's going off memory, and my memory is pretty shitty.

LoveIsOwlYouNeed · January 2016

Also consider that every month, new couples are added to the mix. So if you literally took 5 couples and said one of them will get pregnant this month; it's not like the next month there are four couples, then three, then two, then one and so you're guaranteed to be pregnant by the end of that five months. It's like you remove the couple that got pregnant and another couple takes their place, so you still have a 1 in 5 chance. Every month.

skilouise · January 2016

Maybe this helps:

Scenario 1:

Month 1: TTC, unsuccessful
Month 2: TTC

Scenario 2:

Month 1: no sex in FW
Month 2: TTC

As far as your body is concerned, once you're in month 2, it doesn't matter what happened in month 1 - whether you TTCed or not. As long as you weren't on hormonal BC, your body is in the same position in month 2, so your odds for that month are 20%.

However, your cumulative odds for the 2 months are higher in Scenario 1, because you had two months of TTC instead of just 1.

mamadomino · January 2016

MrsDho11 said:

mamadomino said:

Does anyone know what goes in to the "20% chance with perfect timing" thing? Is that once in the fertile window? Right before o? I always have so many questions about what that means.

I can't put my finger on the exact origin of that right now, but I believe that it is roughly "if you had sex every 2 to 3 days during your FW".

But that's going off memory, and my memory is pretty shitty.

Good enough for me. Thanks!

Lulucooks · January 2016

If it did compound I'd have a 380% chance to get pregnant this cycle..... You know what that means!!!!

TRIPLETS!

Megzb510 · January 2016

Lulucooks said:
If it did compound I'd have a 380% chance to get pregnant this cycle..... You know what that means!!!!

TRIPLETS!

We will need a cat to touch your belly to confirm. If you have a gurgling uterus, they are identical twins!

Lulucooks · January 2016

Megzb510 said:

Lulucooks said:
If it did compound I'd have a 380% chance to get pregnant this cycle..... You know what that means!!!!

TRIPLETS!

We will need a cat to touch your belly to confirm. If you have a gurgling uterus, they are identical twins!

Luckily I adopted a cat for just such occasion. She said "Try Again Later"..... my fault for getting a black cat. They are the color of 8 balls.

MrsDho11 · January 2016

simcal18 said:

MrsDho11 said:

I will not nitpick. I will not nitpick.

Nitpick away @MrsDho11! Would love to hear from someone who has more experience in statistics than I do. The way I approached it is that if a couple has a 20% chance of becoming pregnant each cycle, there is an 80% chance that they will not become pregnant. So to calculate the probability that the couple still won't be pregnant after 2 months (assuming good fertility and assuming all other variables stay steady, which is rarely true): .8 * .8 = 64% probability of no pregnancy after 2 months. For three months, .8^3 is about 49%, for month 6 .8^6 is around 26%, and so on and so forth. But maybe I'm missing something.

I mean, it's close enough, that's why I said "nitpick".

It's a simple model, but it's useful and good enough for our purposes. There's no need to let me make this unnecessarily complicated.

I wrote out a longer answer including the ways that it is more complicated in that, but I feel like as soon as I started bringing Bayes into it, it became clear that I was taking this too far. Then I said this to myself:

MrsDho11 · January 2016

Also, it's super difficult to find actual statistics on this rather than just articles by sites like this and baby center and webMD that all say the same thing but don't cite their resources (#petpeeve)!

Then I was reminded of the actual term for this, fecundability, and now I have WAY TOO MANY TABS open!

khbear117 · January 2016

NEVER make yourself a statistic. Recent research says 39.6% of men and women will be diagnosed with cancer at some point in their lives. That doesn't make a hill of beans to the person who gets it. For them, they got cancer, end of discussion. My favorite one is being struck by lightning. You have a 1/12000 odds of being struck by lighting. The person who gets struck by lighting still has a really crappy day, even if they have low chances of it happening.

Also, statistics are an average of a large sample population. Not every couple is trying to get pregnant. Not every couple is temping and having sex in their fertile window to increase chances of conception. Not every couple has the same genetic make up. Not every couple has the same health picture. Statistics help when looking at the whole picture, but are not as useful when dealing with your own unique situation.

Also, what everyone else said, it doesn't compound, it's each unique month.

PrettyPalomino · January 2016

simcal18 said:

MrsDho11 said:

I will not nitpick. I will not nitpick.

Nitpick away @MrsDho11! Would love to hear from someone who has more experience in statistics than I do. The way I approached it is that if a couple has a 20% chance of becoming pregnant each cycle, there is an 80% chance that they will not become pregnant. So to calculate the probability that the couple still won't be pregnant after 2 months (assuming good fertility and assuming all other variables stay steady, which is rarely true): .8 * .8 = 64% probability of no pregnancy after 2 months. For three months, .8^3 is about 49%, for month 6 .8^6 is around 26%, and so on and so forth. But maybe I'm missing something.

I think I see what you did there! Is that like a null hypothesis? Okay now I'm more confused!

PrettyPalomino · January 2016

@mrsdho11 Iwillnotgoooglefecundability Iwillnotgoooglefecundability Iwillnotgoooglefecundability

MrsDho11 · January 2016

@PrettyPalomino - nope, @simcal18 is not offering a null hypothesis. If it helps think of "heads" as being pregnant and "tails" as being not pregnant. Now you have a coin that's weighted where the odds of it coming up heads = 20%, tails = 80%. The chances of six consecutive "tails" in that situation is roughly 26%.

mamadomino · January 2016

What I have learned from this thread is that I am twice as likely to be diagnosed with cancer in my lifetime than I am to get pregnant this month.
WHAT. THE. FUCK.

pretzel2plus1 · January 2016

Oooh, I was too late to this thread! Are you a fellow stats major, @MrsDho11?? I love stats discussions!

I went the Bayes route, too, before I read your response.

_darth · January 2016

MrsDho11 said:

simcal18 said:

MrsDho11 said:

I will not nitpick. I will not nitpick.

Nitpick away @MrsDho11! Would love to hear from someone who has more experience in statistics than I do. The way I approached it is that if a couple has a 20% chance of becoming pregnant each cycle, there is an 80% chance that they will not become pregnant. So to calculate the probability that the couple still won't be pregnant after 2 months (assuming good fertility and assuming all other variables stay steady, which is rarely true): .8 * .8 = 64% probability of no pregnancy after 2 months. For three months, .8^3 is about 49%, for month 6 .8^6 is around 26%, and so on and so forth. But maybe I'm missing something.

I mean, it's close enough, that's why I said "nitpick".

It's a simple model, but it's useful and good enough for our purposes. There's no need to let me make this unnecessarily complicated.

I wrote out a longer answer including the ways that it is more complicated in that, but I feel like as soon as I started bringing Bayes into it, it became clear that I was taking this too far. Then I said this to myself:

Hahahah. I popped in to bring Bayes' rule into it. I'm gonna just pop back out. @MrsDho11 has got this.

MrsDho11 · January 2016

@pretzel2plus1 - nope engineering!

MrsDho11 · January 2016

And awww @DarthFuriosa! I was wondering all day if you were going to come play with me!

g8rmama_16 · January 2016

I literally failed stats in college. Like, I got an F because I didn't drop it in time. Everything about this post is giving me major anxiety.

Luckily I aced it the second time around.

catiecatp · January 2016

As someone who teaches stats to psychology undergraduate students... I must say I love the level of nerdiness this thread has. Things like null hypotheses and laws of large numbers make my heart go pitter patter.

Want a personalized experience?

20-25% chance each month - does this compound??

Re: 20-25% chance each month - does this compound??

Choose Another Board

Search Boards