Status:  Tried all the basic stuff - bonding - staking.  I'm now playing around with the code and exercising all the features to best understand how to manage this asset as a part of my portfolio.  Finally found a good OHM analytics page thanks to Dune

Olympus intro videos:

Note: these videos are definitely dated.  Olympus changes almost every day with new features, etc.

Olympus intro video 1 (What is OlympusDAO) paraphrased (short video):

I expand a bit to fill in context here:

• Olympus is a dapp.
• The dapp is a set of smart contracts that seeks to keep a cryptocurrency (OHM) pegged (currently) to DAI, which is pegged (currently) to USD.
• If OHM, which is traded on the open market, is valued higher than one DAI, then the dapp (hereafter the "treasury") prints OHM and dumps it on the market in an effort to suppress the price. (in actuality, they don't care if it has a higher price than 1 USD, just that it never drops below)
• If OHM is trading lower than one DAI, then the treasury buys it on the open market and burns it to reduce supply in an effort to raise the value relative to DAI.
• Treasury sells bonds in exchange for crypto assets that it holds in the treasury
• Buyer of bonds gets Ohm at a discount on bond maturity 
• It occurs to me that they use a lot of defi legos like DAI and Uniswap/sushi to achieve a practical outcome - they didn't have to figure out and implement stablecoins, liquidity pools, and AMMs - the hard parts are already primitives (not taking anything away from Olympus - it looks clever - but it would have been extremely hard without these defi legos that took many years to make).  Some governance by DAO - need to explore how much.  Basically, olympus is a smart contract that sells bonds against, and holds assets securing, a cryptocurrency.

Olympus intro video 2 (How to get OHM and stake it) paraphrased (short):

You really should read this first, but don't have to: https://olympusdao.medium.com/a-primer-on-oly-bonds-9763f125c124

• Buy some the DAI OHM liquidity pool on sushi and stake it on olympus.  That's about it. (Note: you can just buy OHM and stake it if you don't want to buy a bond)

Olympus intro video 3 (Olympus Game Theory) paraphrased (short video):

• Nice definition of game theory, then statement that all people buying/selling/staking OHM and OHM bonds benefit from the success of the protocol - if everyone sells out of the system, they lose money - fair enough - not surprising.  It so happens that in the Olympus system, staking (and holding) is the most profitable for everyone if everyone does it.
• Not in the video, but from medium posts - the stakers split the treasury rewards for any given unit of time - if you are the only staker - you are basically getting all of the staking rewards - so it is very unlikely that there will ever be no stakers as long as there is a treasury.
• What is not mentioned is the financial system outside of Olympus, DeFi, tradfi, regulatory systems, etc.  closed system analysis.

Olympus intro video 4 (Sherpa academy | OLY 101: Basics of Olympus) paraphrased (1 hour 45 minutes):

Recommended reading to prepare for this video (I'll weave statements/facts/derivations from these into my comments on the video):

A primer on bonding: https://olympusdao.medium.com/a-primer-on-oly-bonds-9763f125c124

There are some unexplained terms right at this stage - I believe they are referring to OLY as the premium for an olympus bond unit over the value of DAI?  They don't really say - probably was explained elsewhere.

Dai Bonds: A more effective sales mechanism: https://olympusdao.medium.com/dai-bonds-a-more-effective-sales-mechanism-c9a57586f1f7

Notes on the video:

• This is the important video in the series.
• Interesting mention up front that USD is not stable, therefore DAI is not stable, so lets create something else (Andreas would say that's Bitcoin) - and a quick mention of pegging to a basket of assets and growing a treasury-type currency to a large size in an effort to seek price stability (maybe that's a future idea - I know DAI is still planning on this)
• 1USD (via DAI) per OHM is the minumum value target. The protocol/treasury/dapp always values OHM at 1USD and seeks that value. 
• When Olumpus mints OHM, it manifests value to meet demand for OHM, and it shares that manifested value with stakers.
• Interesting that "Risk" in risk-free-value refers to risk of deviation in value from USD - ETH is viewed as more risky than DAI/USD as it fluctuates against USD, and could go to zero USD - the fact that USD could go to zero against everything is not considered.  I mean, I'm fine with it, but that's the definition of risk free value for the purposes of the Olympus team - risk of deviation from USD - really different from the opening statement. 
• Three bond types (not sure these are all implemented - I'll have to try them)
• LP Bonds - liquidity provider bond - they want you to create liquidity on Sushi that they can buy and hold - traders pay Olympus the usual fees as Olympus is holding that liquidity on sushi allowing traders to make trades (currently thats some insane amount of income into the treasury per day loaning liquidity - if you sell OHM, the treasury gets a cut from the liquidity fees, aren't the AMMs cool as nucking futs?)
• you deposit cryptoassets
• you get a bond redeemable for a premium in OHM
• Deposit is in the form of LP-DAI pool from sushi.
• the protocol actually "splits" the profit of a bond sale with the LP bond buyer.  the buyer gets OHM below market, and the protocol gets more than the risk-free value it sold in OHM.
• Contention is made that bonded OHM goes up in value quite rapidly in value as the liquidity pool grows in size - math not presented at this point - I think this is covered later 
• "Stablecoin" Asset Bonds 
• Not much said - these seem to be OHM bonds for DAI. 
• "Unbacked" Asset Bonds
• They are accepting ETH for OHM bonds as well
• They currently have enough assest in the treasury to fund a reasonable APY (351 days at Sep 17th APY) for stakers for about a year if no more bonds were purchased - something like 27 USD in value per OHM. 
• The fact that the protocol provides it's own liquidity really makes sure this is always tradable and profitable - this is super important.
• Staking
• You are putting OHM in a pool 
• Pool gets some minted rewards every 8 hours (currently .4% every 8 hours) (this is why they are saying you get exponential growth, you own a percent of a pool that grows a certain percentage every 8 hours - but it's a weird equation - anyway your shit compounds - this is a really high APY as it compounds (if you let it))
• The DAO sets the APY, which will go down over time - this is treasury management
• It makes no sense not to stake OHM during the bootstrap phase (now) if you have it - agreed - there is no other way to make more on this thing - you encourage price appreciation and make compound interest at ridiculous rates. 
• Treasury as a service idea
• Since they have 27 USD in value in the treasury curently for every OHM, they can , if they want to, gamble with 26
• They currently put some in dai/aave, frax/convex, collecting yeild for the treasury
• Fun puzzles at the end
• Olympus 201 coming soon.  These are great.  I will be in that one live. 

Why is Olympus Useful to study?:

They are exploring improvements to treasury management using cryptoassets - treasury "control" was popularized during the development of Olympus, and Olympus might be the most prominent project to emphasize that.   If you are into that, this is a good study.  It is not useful to study as an investment unless your strategy is to invest in lots of crazily risky things (not that there is anything wrong with that).  Olympus is also often accused of being a ponzi scheme, even though it is not - mistakes may have been made in outbound communication and in defining cultural norms that really hurt Olympus that others can learn from.  The core dev team and investment team have odd demographics as well - there are VCs, anons, other project teams, and strong arguments to be made that decentralization of control of Olympus is poor.

Mechanics of using Olympus:

There is nothing complicated about it - TL;DR is trade on sushi for OHM, and then go to the olympus website to stake. 

Calculator guy has a nice video on staking v. bonding: https://www.youtube.com/watch?v=nIwDU9L9KYU (and one on using your staked OHM tokens to take more risk on by lending and re-staking them: https://www.youtube.com/watch?v=9hTe4BaX8fs.  There are zillions of videos out there on more complex DeFi trades, but they all seem to increase risk higher than reward - really hard to calculate - more on this later.

Olympus Pro:

Really interesting strategy of Olympus to be an LP for lots of other projects as a service (taking 3%).  Justin Bram has a good take on this.  https://youtu.be/WqjbWVjYMHg?t=185.  Nat Eliason covered Olympus Pro in a pretty good article on OHM as well: https://every.to/almanack/olympus-dao

33 Together:

It appears that the treasury supports a prize drawing for stakers.  Wild.  I completely cracked up when I saw that - did not expect it - just a really interesting thing - maybe national treasuries should do that.  The way it works is that everyone who wants to participate forgoes staking rewards so that they can be put into a prize pool, which you have a chance of winning commesurate with the staking rewards you gave up - minus a tiny fee for development, I think.  Sort of a 1:1 fun gambling thingy.

Team/Dao:

Listen to the podcasts and join the discord to get a feel for them.  I think it would be best to attend their online events as well.

Dao voting seems pretty strongly unified - but a LOT of things are passing that seem like they will be a lot of work to implement - not sure how they will be able to focus when they are committing to so much. 

There were a few VCs involved early on, and 11.8% of the OHM is reserved for the team and VCs to release after certain requirements are met - it seems like a reasonable mechanism overall, but very little is said about it - I don't fully understand the vesting schedule - probably a TODO for me, and it really should be a todo for the team to make that clearer.

Economics/Investing:

Note: ATM, I'm still checking some issues with my logic, so I'm almost guaranteed to be at least slightly wrong about something, but I don't think I'm meaningfully wrong for the purposes of this section.

TODO: I don't have a great handle on circulating supply growth, other than from this Dune Analytics Page

OHM trades in a wide, generally up-ticking range of about 10x since it's release( ATL/ATH 162/1415 as of this writing ).  However, even if you buy in at the top of that range, and hold for a year, at, say, 8000% APY, which the Treasury has enough reserves to cover for over a year (AFAICT), then you are not too worried if you sell at the bottom of the trading range (80X units of currency at 1/10 the price is just fine for a yearly investment). 

The APY is a bit hard to pin down, it' based on holding for 365 days at the current APR, which changes every 8 hours.  It seems to have started at an astounding rate and leveled off at about 8000 recently.  What you actually get is an 8-hour rebase rate, which gives you an APR, but because you are compounding every 8 hours, that yeilds a very high APY.

Guessing at APR and price: Voting is involved in setting the APR, but I haven't participated in that, yet.  I'll fill in some info after I do.  Really hard to say where that will put the average APY or effective APY for long term holders this year.  It seems to be expected that the APR will decrease continually down to a reasonable interest rate (10%?) over the course of the first year.  In that case, the price of the OHM token should drop to just above the Risk Free Value at sale.  Some math needs to be done, here, and I think we might be out of scope for google sheets.

Risks are the usual - this is a new token and it could certainly drop meaningfully below the current trading range (say, 1/80th of the current price - giving yearlong holders a loss)  when you need to sell for whatever reason - that is probably the biggest risk - contract risk being the other biggie. 

The most likely thing that will happen in a real bear market for OHM is that the price will drop to just above the treasury reserve amount per ohm (the backing price), which today is about 1/10th the cost of 1 OHM (again, AFAICT - I'm trusting the numbers on the OHM app for the moment), so if you haven't made 1000% (a few months at this rate) on your OHM by that point and need to sell, you're going to sell at a loss.  For long term holders that seems a reasonable thing.

There is the issue of truly understanding OHM supply growth, and how the DAO strategically will adjust rebase rate and thereby APR/APY.  It seems like printing is algorithmic, and interest (rebase rate) is DAO controlled, but I just don't have those nailed down, yet. 

Aside: I guess a partially DAO-controlled treasury is almost inevitable at this stage - even totally algorithmic systems like Bitcoin are kind-of DAO controlled by core dev updates - humans still turn on and off the computers, for the most part - maybe that's a good thing.

Lots more to add here.  Basically, contract risk aside, you probably want to make sure that the treasury has enough assets to give you an acceptable return over your investment timeframe in the worst case scenario before you invest.  This is a high-risk investment, but a very interesting experiment in treasury and reserve currency management.  Like almost every other cryptoasset worth investing in - getting in early is helpful - doing the math is more helpful.

The docs have an example of a treasury run scenario here.  And there is a pinned DD post on reddit

NOTE on TIMING: Just looking at the chart - it looks like whales are potentially swimming in the OHM pool, as we see some aperiodic price volatility - general trend is up, but you might consider the history of the chart, and dollar cost average in over a period  that gives you good confidence in your average purchase price / tax basis.  Also, OHM does major contract upgrades from time to time, and some significant volatility has been reported around those - perhaps they take a while, and are viewed as high risk events, or present opportunities for individuals and exchanges to f-up one thing or another - something to consider when buying in.

Code:

All on Github: Treasury.sol and Staking.sol are important.  

Importantly, I can easily find, read, and understand the better part of these contracts, and the developer documentation.  I can't say that for Wonderland.  This is really a critical step in building the confidence to invest in the project.

TODO: I need to step through the supply regulation code - it's not immediately obvious how the Treasury regulates supply. 

Olympus - the Big Idea:

A good friend asked me about the big idea behind Olympus DAO.  I'm just pasting in a bit of the conversation we had for now:

> I was thinking about this a bit more...let's say the goal is to build a viable alternative that eventually kills off Wall St. / Stock Market

The closest understanding I have come to of a common goal for Olympus contributors is that they are looking to do two things:

First, create a stable currency that is not dependent on any FIAT currency, and, in fact, replaces FIAT stablecoins for many purposes.  I suspect that most people think that transparent, predictable treasury management is important.  It might take them a very long time to accumulate enough assets of enough types that OHM is more stable than USD - this is sort of the "development" phase and they are offering crazy interest rates to OHM investors while they build that capability.

That's the first part. They have developed what is, to me, a fairly complex system to make that work.  Reading about it feels the same way that reading about Steemit felt to me - id doesn't feel like it needed to be this complex to make it work - but it does make sense that it works and it seems brilliant.  You can buy their OHM bonds with your assets, or stake OHM and get staking rewards - and there are a couple other things I haven't tried.  Given the goal I identified - all of it is just a way to fill their treasury with assets that they think will make the token stable.

Second, they decided to create a permissionless system whereby any DeFi project can sell "bonds" as opposed to ICOs or Liquidity mining.  They wrote an article on that here.  Obviously, I haven't tried it - but it would give projects another fundraising alternative that allows them to build their own treasury over time.

That second product is called Olympus Pro, and one way it works is by using LPs formed of the Olympus native token and the token of the DeFi project.  I'm fairly sure that if other DeFi projects like that product a lot, then it makes OHM useful as something other than a stablecoin after the growth phase.  

>Which, I like that goal in so much as it should theoretically help more evenly distribute wealth and access to equal investments.

> So how does DeFi 2.0 do that? Just providing an alternative is not enough right...actually getting a critical mass to prioritize DeFi over traditional markets leads to real success.

> And so how do you that? Well, seems like the way to do it is you make people rich from betting on DeFI.

I think this is one of those cases where Olympus is picking off a couple bits of traditional finance - building better-run treasuries and more liquidity options for startups.  You are right, though, that during the growth phase of Olympus, they are offering ridiculous interest rates to stakers, which is why I am making big (for me) bets on OHM and KLIMA right now. 

> Which, really, I think is what convinced a lot of people to put their money in the markets in the 20's and then again post-depression. I remember reading a good book about "bucket

> shops", mom-and-pop gambling on the market ( https://en.wikipedia.org/wiki/Bucket_shop_(stock_market) ) and how that generated a lot of common person interest in the markets.

Those are fascinating.

> So I guess that is kind of what is happening in DeFi right now in a sense of people kind of setting up bucket shops and some people are getting rich off of it. I just hope that the wealth

> gains are distributed as widely as possible rather than just the rich getting richer.

I think the larger share of DeFi profits going to nerds like us to redistribute as we see fit, or keep.  Some speculators and finance wonks are making bets, too, but I believe they are less able to identify good projects and less able to secure their systems - still, I feel the DeFi investment markets are much less rigged than the CFi markets, so there is a little more breadth of opportunity.

> Which leads me back to what I've been thinking a lot about lately, which is essentially if we treat wealth in the U.S. as a non-zero sum game, where can we most efficiently pull wealth > out of some places and put it into other places which will increase overall wealth fastest or at its pareto-optimal rate. I'm not sure an unrealized capital gains tax does that (though I'm > open to being convinced otherwise). Could DeFI do that? I guess it depends on what everyone gaining newfound wealth with DeFi is putting their wealth into, especially in what

> capital-C Capital it ends up ultimately financing.

> What do you guys think?

I do, currently, think that DeFi is where wealth redistribution experiments will take place.  It just moves way faster and there are already some great experiments going on - projects like Circles are learning a lot, and they are nowhere near as constrained as national governments are in running experiments.  Further, once an experiment is deemed interesting, it will just get forked and tweaked freely by many other interested parties.  Software is eating government.

Other notes:

See Also:

Justin Bram has some really good videos related to OHM and derivatives - totally worth checking out his youtube channel and the Defi Innovation discord.

Reserve or this podcast on reserve - not quite as interesting as OHM.

Wonderland (on avalanche) - much harder to gain confidence in - heavily MIM dependent - not a lot of dev docs - harder to talk to the team and to find the code.  My understanding is that TIME is partially backed by OHM (a bunch of OHM bonds in the treasury, perhaps) but it's not immediately clear to me what the treasury strategy or DAO governance is.

From Olympus DAO reddit:

Not sure the rate going dow is a guarantee. There will be a vote by Ohmie community in January/February, which is when OHM supply should reach a certain circulating supply milestone of 10mil. Right now the 8kAPY has a 314 day runway, so I would vote to keep APY the same if the 8kAPY runway is above 300 days in February.

OHM distribution: https://dune.xyz/CyJackX/Olympus-Wallets