cross-posted from: https://lemmy.world/post/1362188

This is an analysis of the available number of shares through different filings and publicly available information. There are no shares available unless they are borrowed from an institution or multiple institutions sell their entire positions. The only thing is that borrowing more shares will achieve nothing since the shares need to be sold by apes or there will be an even bigger problem since apes will probably be buying. There is no escape for any entity short GME, except maybe the first to close.

First let’s establish the number of shares in total and directly registered:

From the 2023 10-Q, there are approximately 304.7mm shares in total, and there are approximately 228.1mm shares not directly registered by apes.

We will start with the number of shares not directly registered by apes, approximately 228.1mm shares. First, we can remove Ryan Cohen’s shares, 36,847,842 shares. Next, we will remove the known institutional holders, this amounts to 86,113,117 shares. The total of these two groups is 122,960,959 shares.

228,100,000 - 122,960,959 = 105,139,041

There are only 105.1mm shares that are not owned by RC, institutions, or directly registered to apes. Many of these do not actually possess their shares though as about half the institutional shares are publicly loaned out… While almost half of all the loans are from Vanguard as almost their entire holdings of 25mm shares are loaned.

Now, to estimate the minimum amount of non-directly registered retail shares, we can take away the retail broker non-votes disclosed in the last 8-K:

There are 60.7mm broker non-votes.

105,139,041 - 60,764,147 = 44,374,894

This leaves 44.37mm shares that are available. That’s 14% of the total shares and the publicly reported short interest is 55.16mm shares or 18% of the total:

There are 55.16mm shares publicly reported short.

44,374,894 - 55,160,000 = -10,785,106

This means that the maximum number of available shares is -10.78mm shares. Yes, that is negative.

Again, this is only using the minimum number of shares of each group. There is obviously more retail than the broker non-votes. There are two things that I wonder about the yearly vote. Did any institutions or funds vote? If not, that would mean there is about 20mm more shares for minimum retail. This does not include any retail whose broker does not allow voting and provides no broker non-vote. If institutions did vote, why are the vote tallies much less than the total shares? Are the results normalized for the non-institutions (retail and insiders)? If so, then shorts are even more impossibly screwed… If we assume that there are (mostly) only non-normalized retail votes, then we can create a range within which the maximum actually lies. There were 129,496,196 votes cast. We can remove the directly registered shares and RC’s assuming he voted (lol).

129,496,196 - 76,600,000 - 36,847,842 = 16,048,354

-10,785,106 - 16,048,354 = -26,833,460 so maximum range of [-26,833,460, -10,785,105]

76,600,000 + 60,764,147 + 16,048,354 = 153,412,501

With retail owning 153,412,501 shares. That’s nearly double the institutions. Minimally there are at least the DRS count plus the non-votes which is like 136.8mm shares.

I do not see how millions of shares could be trading every day. Institutions, mutual funds, and insiders have been increasing their holdings. Around 1% of the total shares are being traded daily but around -5% is maximumly available, unless retail is churning 1% everyday… However, that does not make any sense because then the short interest would be decreasing as shorts covered from retail selling. But they cannot cover, this is known if apes are not selling and proven by more than two years of the same publicly reported short interest. I think that there are multiple market participants that have overleveraged swaps with these institutions (probably Vanguard (see 1. B. iii)), and they are trading back and forth using the swaps to try to reduce their maintenance burden. Ya know, just the thing that Bill Hwang/Archegos was doing before blowing up… That is the only explanation that I could possibly conceive and explains why the CFTC is so adamant on non-reporting swaps.

In order for the shorts to escape they would have to negotiate with several institutions to sell their shares, but why would the institutions do that if they could sell them a small fraction of those shares for the same (or larger) dollar amount when the shorts blow up? Lending more shares just means that they won’t be able to sell the shares themselves when the shorts inevitably blow up…

And we’re dumb money, lol.

Sources:

Form 10-Q for Gamestop Corp filed 06/07/2023

Form 8-K for Gamestop Corp filed 06/21/2023

SEC.gov | EDGAR | Company Filings

GameStop Corp. (GME) Stock Price, News, Quote & History - Yahoo Finance

GameStop Corporation Common Stock (GME) Stock Price, Quote, News & History | Nasdaq

GME (NYSE) Gamestop Corporation - Class A Stock - Short Interest, Short Squeeze, Borrow Rates (fintel.io)

Posted via @Zuberi Reddit-scrapper (Manual Reposts). PM me for suggestions.