AMD’s stock plunges to biggest loss in over 12 years | Hacker News Reader

AMD’s stock plunges to biggest loss in over 12 years

marketwatch.com

27 pointsbaazaar26 comments

A clickbait headline if you have ever seen one. Sure, it's the biggest percentage loss, but then of course AMD rose four fold in the last year or so.

AMD on May 3, 2016: 3.7

AMD on May 1, 2017: 13.6

AMD on close May 2, 2017: 10.32

It's still more than double, this is just the natural correction after all the hype. And since they went 4x in a period of 1 year where .. nothing .. really happened (until Ryzen very very recently) there was a lot of fluff.

I honestly regret not buying some AMD stock a few years back, even though I was confident that it wouldn't be abandoned after the Mubadala acquisition. Just like I regret not buying BTC when it was I first heard about it (~$6/BTC). Oh well haha :P

Well, obviously you remember the times you passed on buying something that turned out to be worth a lot... there are probably lots of times you passed on buying something that turned out to be worth nothing.

Definitely! I'm completely fine with it though. It's not something I lose sleep over, just a little itch is all.

Riddle: if I purchased on May 3 2016 I'd be worse off selling on May 1 2017 than holding. Why?

Long term capital gains vs short term?

Doesn't compute. You're only taxed on the _gain_, which means as long as your selling price is greater than your purchase price then, ignoring inflation, commissions, etc, you're always ahead as long as the tax rate is less than 100%.

Anyhow, the closing price on May 3, 2016 was $3.60. The closing price on May 1, 2016 was $13.30. Even if the entire sale price was taxed, and assuming you're already fairly wealthy, you'd need a marginal tax rate of well over 50%. The top marginal federal income tax rate is %39.6; the top state marginal income tax is 13.3%. Even a 53% tax on the total sale price wouldn't cause you to lose money.

While true, you are taxed at different rates for different holding periods. If you hold for < 1 year, that is a short term capital gain, and is taxed at your ordinary income rate (~35%, let's say). If you hold for > 1 year, your gain is taxed at 15%.

Sure. But there's always a gain, regardless. The riddle was why would you be worse off holding rather than selling.

Anyhow, if we grant the riddle was poorly worded, that the implication was waiting an additional day you're be better off, it's still technically incorrect. The clock starts ticking the day after the purchase, and you have to hold for a year and a day. If you purchase on May 3 2016 you shouldn't sell before May 5 2017. Possibly May 4 or even May 3 if I'm misunderstanding something, but almost certainly not May 2.

> Sure. But there's always a gain, regardless.

Not true. Suppose you buy a stock for $5 on Jan 1 2016. It rises to $20 on Jan 1 2016. If you sell on Jan 1, you make 15x.65 9.75. If you hold until Jan 2 and the price declines by $1 to $19, you make 14x.85=11.9.

Hence, you're better off holding till the next day, even though the stock goes down.

In this example, you are better off taking the short term capital gains penalty (I did the math in a sibling comment).

However, your logic isn't correct; no one is arguing that you would be better off having never purchased the stock at all versus paying short term capital gains. What we are trying to figure out is if it is worth taking LESS profit in order to pay LESS taxes. This calculation depends on how much less profit we are talking; as long as the profit lost by waiting until one year for long term capital gains to kick in is less than the difference in tax rate, it is worth it to wait.

You can prove that both situations are possible is easy to show. If your profit before long term kicks in is $10 and the profit after is $9.99, it is obviously better to wait until long term kicks in; you are going to pay WAY more than 1 penny per share in extra taxes. However, if the profit goes from $10 to $2 a share, it is obviously better to take the higher profit and pay the extra tax.

Although, in doing the math, I still think you would be better off taking the short term capital gain hit for the extra profit - if long term is taxed at 15%, and lets say you get even 40% hit for short term, you are still better off taking the higher share price:

(13.6 - 3.7) * 0.6 = $5.94 profit per share (10.32 - 3.7) * 0.85 = $5.63 profit per share

Should sell and take the short term capital gains hit!

It's still 25% in a day. Yes, it shows disappointment and large correction from all the hype. Not that clickbaity.

Definitely not clickbait, this is a huge movement for any stock and it happened due to genuinely disappointing news from AMD

Correct me if I'm wrong here, but this is saying "their stock tanked, because they had/are having a 'not perfect' quarter, and we based their value on them having a perfect quarter".

That seems like they only have themselves to blame-to expect anyone to ever have a 'perfect' quarter and to think nothing will go wrong at all, (and to write off a quarter that is not-perfect-but-still-good) and to base all your expectations on that seems incredibly, incredibly stupid.

Is that what's happening here? Or am I missing something?

Sounds like classical Planning Fallacy to me.

https://en.wikipedia.org/wiki/Planning_fallacy

In a nutshell, the theory is that people tend to plan too optimistically because they look at the (usually small) probability of each problem, but fail to consider the much larger probability of there being problems at all.

..... After having gone up 400% in a year

I'm tempted to buy some AMD now after this gigantic one day plummet. While I'm not sure AMD will ever be able to compete with Intel in the enterprise cpu market (cloud providers), NVIDIA and GPUs are looking like a solid core business for them.

Everybody keeps on hyping and saying Ryzen, but again... I am very skeptical major enterprises will make the switch from Intel.

(I have dyslexia, and originally read different companies.) Dont do it. It would be a risky, short-term strat. NVIDIA is the future, not AMD/ATI, at least right now. The move seems like an emotional market move as a reaction to institutional-investor liquidity event in-touch with future value. AI is going to be a trillion dollar business and will consume a huge swath of Tesla GPUs, and will need zillions of them to deliver pervasive value at scale. That is, before specialized AI accelerator ASIC and FPGA (combined with improved energy efficiency across the stack (ram, compute, networking and storage) and many, many cores) solutions are optimized to the various problem domains... then exit NVIDIA before that happens. AMD/ATI would be wise to pivot into going all in on AI acceleration, and that would be possibly GPU-killing for that market if GPU product evolution doesn't keep up (doubtful it would happen but at least AMD would have a stronger chance of survival.)

What industries is AI expected to be so useful in, and what for?

Serious question. I'm interested to learn more.

I think the main one that people are banking on is self driving cars. Whether it'll get here in 2 years, or 10 years. As soon as it's production ready, entire industries will be replaced

Are there any other industries where AI is expected to be big?

I'm pretty bearish on self-driving cars.

So, now's the time to buy?

May 3 2016 would have been better

Rolland's advice was neutral (neither buy nor sell) but his price target was higher than it is currently, so it might be good to buy. On the other hand, Chin's rating is "sell" and his target price is quite a bit below where it is now. I guess it depends on how you think they will do with Vega.

Yesterday might have been the time to short it

Reply on news.ycombinator.com