# How to Spot 10x Vertical Spreads BEFORE They Take Off

https://www.youtube.com/watch?v=6I5a3QQX4y0

[00:00] Do you want to trade vertical spreads with 5x or even 10x returns?
[00:04] Well, in this video, I'll walk you through exactly how I find these trades before they take off using volatility skew and momentum.
[00:12] I'll also show you trades we placed that returned multiple hundreds of percent in just the past month, so you can start trading them, too.
[00:18] You might be wondering, what does volatility skew have to do with vertical spreads?
[00:23] The short answer, a lot.
[00:25] But to fully appreciate its impact, let's first understand what skew is, where it comes from, and why it matters in practice.
[00:32] The Black Scholes merin model foundational in options pricing theory makes a critical assumption.
[00:38] Implied volatility is constant across all strikes and expirations.
[00:42] In other words, the model assumes that no matter which strike price you choose, the expected volatility of the underlying asset remains the same.
[00:50] This assumption stems from the idea that the underlying follows a logn normal distribution meaning that log returns are normally distributed.
[00:58] But reality disagrees.
[01:00] Empirical studies across markets including equities have consistently
[01:01] Shown asset returns distributions are not perfectly normal.
[01:06] Instead they tend to exhibit negative skewness asymmetry with heavier left tails.
[01:10] That means large drops are more likely than rallies and excess kurtosis also known as leptokurtosis.
[01:15] Fatter tails and a sharper peak than a normal distribution.
[01:19] This means extreme outcomes, especially crashes, are more common than the Black-Scholes model would suggest.
[01:25] Let's take SPY, the ETF tracking the S&P 500 as a real world example.
[01:31] If you plot the historical distribution of SPY's log returns, and compare it to a normal distribution, you'll notice negative skewness.
[01:37] The left tail is longer, which means large downside moves are more frequent.
[01:41] Leptokurtosis.
[01:44] The center of the distribution is more peaked, but the tails are fatter, implying both high frequency of small returns and higher chance of extreme returns on either side.
[01:53] The market doesn't just blindly use the Black-Scholes model.
[01:55] It adapts.
[01:57] And one of the ways it adapts is through implied volatility skew, also referred to as the implied volatility surface.
[02:01] When extended across all strikes and maturities.
[02:03] Skew refers to how implied volatility changes across different strike prices for options with the same expiry.
[02:09] In equity indices like SPX or the ETF SPY, out-of-the-money puts typically trade with a higher implied volatility than at the money or out of the money calls.
[02:16] This produces the characteristic volatility smirk, a downward slope when plotting IV against strike.
[02:23] But why does this exist?
[02:25] Part of the explanation is statistical.
[02:28] The skew reflects the market implied distribution of returns which incorporates the empirical tendencies we just discussed.
[02:34] In quantitative finance, this can be visualized through the risk-neutral density, the distribution of future prices implied by option prices under the risk-neutral measure.
[02:43] Tools like the Breed and Litzenberger formula allow us to extract this distribution from observed option prices.
[02:49] When we do this for SPY, we see that the risk-neutral density shows fatter left tails, mirroring the market's anticipation of downside risks, something the flat Black-Scholes model does not replicate.
[03:01] Skew
[03:04] Also reflects a key empirical relationship.
[03:07] The negative correlation between spot returns and implied volatility in equity markets.
[03:11] When markets fall, volatility tends to spike.
[03:14] This is a well documented phenomenon.
[03:17] This spot volatility correlation is one of the primary drivers of skew.
[03:21] When the underlying falls, for example, SPY drops, implied volatility rises.
[03:26] Therefore, out of the money puts, which become more relevant during market drops, are priced with higher implied V ahead of time.
[03:33] This forward pricing mechanism means that skew is already anticipating a volatility jump if the market sells off.
[03:40] That's why skew is sometimes described using the terms sticky strike where implied V stays the same at a given strike even as the underlying moves.
[03:48] Beyond statistical reality, skew is also a reflection of supply and demand imbalances in the options market.
[03:54] Let's revisit the risk neutral versus real world return distribution.
[03:58] The implied distribution from option prices often overstates the likelihood of downside events relative to historical observations.
[04:03] This leads
[04:05] To what's known as the skewness premium,
[04:07] an empirically observed phenomenon where options with high implied skew, for example, out of the money puts and sometimes out of the money calls are systematically overpriced in volatility terms.
[04:16] This mismatch opens up opportunities and explains why strategies like selling out of the money puts on indices have historically generated positive expectancy.
[04:25] But why does this premium persist?
[04:27] It's largely due to demand for downside protection.
[04:30] Institutional investors with long equity exposure routinely buy puts to hedge tail risk.
[04:34] This persistent demand drives up the prices of implied volatility of these puts.
[04:38] On the supply side, market makers or volatility sellers are only willing to absorb the risk if compensated with a risk premium, meaning higher implied volatility.
[04:49] There's also the lottery ticket effect on the upside.
[04:51] Retail and speculative participants often buy out of the money calls hoping for a big move.
[04:55] These options are cheap in dollar terms but offer asymmetric payoff profiles.
[04:59] Small cost for potentially large return.
[05:04] To take the other side, sellers demand a premium due.
[05:06] To the unattractive payoff.
[05:08] Small premium collected versus large potential loss.
[05:13] In both cases, sellers are offering convexity and they demand to be paid for it.
[05:17] So, how does all of this relate to vertical spreads?
[05:19] Vertical spreads involve buying and selling options at different strikes, which means they're directly exposed to the shape of the volatility skew.
[05:27] The difference in implied volatility between the long and short legs impacts the spread's initial value in Greeks, its P&L profile as the underlying moves, and the edge, positive or negative, you're taking when entering the trade.
[05:39] Understanding skew is not optional.
[05:42] It's essential for any trader working with directional spreads or non at the money positions.
[05:47] It reflects the market's belief, its hedging pressure, and its statistical reality all baked into a single surface.
[05:54] Once we understand that skew reflects both market expectations and structural demand imbalances, the natural next step is this.
[06:00] How can we measure when skew is rich and profit from it?
[06:04] In other words, how can we detect when the skew premium is stretched beyond normal levels, offering?
[06:08] Us a potential edge?
[06:10] Let's break this down into two components.
[06:12] Analyzing the skew relative to its history and comparing realized versus implied volatility at specific strikes.
[06:18] Just like volatility, skew isn't static.
[06:20] It fluctuates over time.
[06:23] By tracking a time series of skew, we can spot mean reverting behavior and identify moments when skew is unusually steep or inverted.
[06:31] Since there is typically a persistent skew premium, especially in equity indices, large deviations from historical norms can often serve as contrarian trading signals.
[06:40] Using the volatility dashboard, let's take a look at a recent setup in open.
[06:44] The skew on the call wing was extremely steep, significantly more negative than its historical average.
[06:51] To quantify this, call skew is measured as at the money implied volatility minus the 25 delta call implied volatility normalized for the at the money implied volatility.
[06:59] A more negative value means out of the money calls are trading with higher implied volatility than at the money calls.
[07:06] This suggests strong demand for calls often from speculative buyers
[07:08] Chasing upside.
[07:10] Likewise, put skew is measured as the at the money implied volatility minus the 25 delta put implied volatility normalized for the at the money volatility.
[07:18] Again, a more negative value indicates that out of the money puts are priced with much higher implied volatility, typically due to institutions seeking downside protection.
[07:27] You'll often see skew spiking during periods of extreme market sentiment.
[07:31] For example, in fear-driven environments, put skew steepens due to aggressive hedging.
[07:35] In greed-driven environments, call skew steepens as traders chase upside via low delta calls.
[07:41] When either skew becomes historically stretched, it creates potential trading opportunities, especially with structures like vertical spreads that allow us to fade overpriced volatility at one strike while hedging it with another.
[07:52] If you're interested in doing this type of analysis yourself, check out Oakquants link in the description and join our community where we share setups like these regularly.
[08:00] In addition to tracking skew levels, it's important to compare what the market is pricing, implied volatility, with what has actually occurred, realized volatility.
[08:05] This is especially critical.
[08:09] At the strikes you're targeting in your trade.
[08:11] As we probably know, realized volatility reflects actual historical movement of the underlying over a specific period, while implied volatility reflects the market's forecast of future volatility and is directly embedded in option prices.
[08:24] To assess whether skew is overpriced, look at the out-of-the-money strike you're considering selling.
[08:29] The wing with elevated IV due to skew.
[08:32] Compare its IV to the realized volatility over the relevant historical period.
[08:36] If the realized volatility is lower than the implied volatility at that strike, it can suggest that the market is overpricing expected movement, creating favorable conditions for selling.
[08:45] At the same time, look at the at-the-money implied volatility.
[08:49] This is usually the option you're buying in a vertical spread.
[08:52] This leads us to our ideal setup.
[08:54] Realized volatility is lower than the implied volatility of the out-of-the-money skewed strike you're selling, meaning it is expensive.
[08:59] And realized volatility is equal to or higher than the implied volatility of the at-the-money strike you're buying, meaning you are buying a cheap option.
[09:07] This dynamic creates a favorable spread.
[09:09] You're buying a fairly priced or cheap option and selling an overpriced one, resulting in a positive expectancy trade.
[09:16] Skew is more than a theoretical curve.
[09:18] It's a real pricing behavior that offers edges to those who can correctly measure it.
[09:22] By tracking skew relative to its historical behavior and comparing realized and implied volatility at relevant strikes, you can identify when the market is mispricing risk due to hedging flows or behavioral extremes.
[09:35] These are the moments when skew becomes actionable that lets you harvest rich premiums while maintaining defined risk.
[09:41] Vertical spreads, buying one option and simultaneously selling another at a different strike but same expiry are often thought of as purely directional bets.
[09:50] But when viewed through the lens of volatility skew, they become something deeper, a relative value trade.
[09:56] The core idea is to buy an option that's fairly priced or underpriced, typically at the money, and sell an option that's overpriced due to elevated skew.
[10:06] This gives you some exposure to the dislocation in the implied volatility surface rather than only betting on direction.
[10:09] Let's break this
[10:11] down.
[10:13] The long leg of the vertical is often placed near the at the money region where there's more liquidity and tighter bid ass spreads.
[10:18] Market participants and market makers are more comfortable providing supply here.
[10:22] So pricing tends to reflect a fairer estimate of expected volatility.
[10:26] The short leg is placed farther out of the money in areas where skew tends to inflate implied volatility.
[10:32] This is where demand outpaces supply.
[10:34] Think crash protection on puts or speculative upside on calls leading to volatility premiums.
[10:40] When skew becomes steep, the relative pricing between the two strikes becomes distorted.
[10:44] By selling the inflated option, and buying the fairer one, you're creating a value spread.
[10:48] Short expensive, long cheap.
[10:51] To find this setup, we want to look for at the money implied volatility that is in line with or below realized volatility over the recent period.
[10:58] Out of the money implied volatility at the skewed wing is meaningfully above both at the money implied volatility and realized volatility, suggesting overpricing.
[11:06] And when these conditions are met, a well-constructed vertical spread can capture the excess skew premium.
[11:09] In steep skew
[11:11] Environments, it's not uncommon to see vertical spreads with reward to risk ratios of 5:1 or even as high as 10:1.
[11:18] To understand how powerful that is, let's reframe the spread as a binary payoff structure and price it using expected value.
[11:26] Suppose a vertical spread pays out a fixed amount if the underlying finishes above the short strike at expiration and expires worthless otherwise.
[11:33] For example, if we can construct a spread with a max profit of $1,000 and a max loss of $100 being the debit paid, the payoff becomes a binary structure where we either win $1,000 or lose $100.
[11:45] A 10:1 reward to risk ratio.
[11:47] Now assume for the sake of illustration that the spread is fairly priced, meaning its expected value is zero.
[11:53] What probability of success does this imply?
[11:55] We can solve for the break even probability P as follows.
[11:58] So the market is implicitly pricing in only about a 9% chance of the spread finishing in the money.
[12:04] Of course, this is a gross oversimplification.
[12:06] Real world option pricing involves a continuous range of outcomes and more.
[12:10] However, this back of the envelope
[12:12] Approach helps illustrate the core idea.
[12:14] If you believe the true probability of success is higher than what's implied by the spread price, then you may have an edge.
[12:21] And if you believe the skewed out of the money option is overpriced relative to the underlying actual behavior, that belief translates into a higher potential realized probability of finishing above the short strike and thus a positive expected value trade.
[12:33] This framework can be useful for intuitively understanding what kind of odds the market is laying on your trade.
[12:40] When skew is extreme and you're selling rich fall, the market is often pricing in very low probabilities of success, creating asymmetrical setups where you only need to be right occasionally to win overall.
[12:52] Let's walk through a real setup using AMC data from July 22nd.
[12:56] We can buy the August 8th expiration 3.5 strike call at the 50 delta for 25 at an implied volatility of 91.3%.
[13:05] And we can sell the August 8th expiration $5 strike call at the 10 delta for 8 at an implied volatility of
[13:12] 145%.
[13:15] The underlying price is around $345.
[13:17] We assume that the past realized volatility around 76% will continue over the life of the spread.
[13:23] While this may seem like a simplification, there's a solid justification for it.
[13:27] AMC's historical variance risk premium, the average difference between its implied volatility and realized volatility, is roughly 18%.
[13:36] That means on average the at the money implied volatility trades about 18% higher than what actually is realized.
[13:43] Now if the current at the money implied volatility is around 91.3%, subtracting the 18% VRP gives us an implied expectation of future realized volatility close to our 76% estimate, which matches the most recent realized figure.
[13:57] So, while we're simplifying by assuming that past realized volatility equals future realized volatility, that estimate is consistent with AMC's long-term IVRV relationship.
[14:06] This makes our volatility assumption both defensible and consistent with empirical market behavior, especially when modeling spread payoff distributions.
[14:13] So, we set up a simulation to simulate this trade using geometric Brownian motion with the forecast volatility based on the realized volatility.
[14:21] And this is what we find.
[14:23] A mean expected return on debit of 23% with a 32% win rate.
[14:29] This implies a positively skewed return distribution.
[14:31] You only win about one-third of the time, but when you do, the payoff is large enough to compensate for the losses.
[14:37] This is a classic positive expectancy structure made possible by selling a highly overpriced out-of-the-money call inflated by skew.
[14:43] Notice how skew is so inflated in this setup that even though we're buying an at-the-money call that is itself overpriced, the entire vertical spread is still profitable.
[14:53] This might seem counterintuitive at first.
[14:56] After all, if you're buying something expensive, how can that be a good trade?
[14:59] Here's how.
[15:01] Yes, in isolation, that call is expensive.
[15:03] But the key is in relative mispricing.
[15:06] The short leg, the 10 delta call at the $5 strike, is trading at an inflated IV of 145%.
[15:12] That distortion more than offsets the modest overpricing.
[15:14] On the long leg.
[15:16] To illustrate this, let's price the vertical under a risk-neutral scenario, assuming a constant volatility of 76% equal to the past realized volatility.
[15:23] Under that assumption, the fair value of the spread is approximately 20.
[15:28] The actual market price is only 17.
[15:31] This difference represents a clear edge of 3 cents per spread or about 18% of capital at risk.
[15:36] In other words, the market is underpricing the vertical relative to its risk-neutral value because the extreme skew inflates the short leg disproportionately.
[15:46] Even though you're paying slightly too much for the at-the-money call, you're collecting far too much premium on the out-of-the-money call enough to tilt the entire trade in your favor.
[15:55] This is a perfect example of why vertical spreads when used with skew are best thought of as relative value trades, not only directional bets.
[16:02] And notably, this result does not rely on forecasting direction at all.
[16:06] The simulation simply generates random price paths consistent with the assumed volatility.
[16:12] The edge comes purely from skew mispricing, not directional.
[16:14] Movement.
[16:16] This challenges the common belief that vertical spreads are only directional tools.
[16:20] When constructed with skew in mind, they can become non-directional relative value strategies.
[16:25] I know we said direction doesn't matter, but let's now take a look at how directional bias affects performance.
[16:30] Using the same spread structure as above, we simulate outcomes under various annualized drift assumptions, which represents directional return trends.
[16:37] Here's how the mean return and win rate shift as we change the drift.
[16:42] This shows that while the trade is profitable even with zero directional drift, adding a modest positive bias significantly boosts both return and probability of success.
[16:51] In fact, it would take a fairly strong negative drift to make this trade meaningfully unprofitable.
[16:57] You'll notice we're still break even even with a strong negative momentum of 50% annualized drift.
[17:02] This leads to a natural question.
[17:04] Can we improve our result by forecasting direction?
[17:06] If direction affects edge and we can forecast direction even imperfectly, then we can compound our advantage.
[17:12] This is where momentum analysis comes into play.
[17:13] On the Oakquans platform, we
[17:15] Implement multiple forms of momentum.
[17:17] Namely, time series momentum measures whether a stock's recent return is positive or negative.
[17:23] Cross-sectional momentum ranks stocks by performance and groups them into deciles.
[17:27] Relative momentum compares a stock's return to its sector or the broader market.
[17:31] And we also offer a few additional signals like turnover, sector momentum, and proximity to 52-week high.
[17:38] Momentum is one of the most studied market phenomena.
[17:41] And these factors have been widely studied in academic literature.
[17:44] Studies have shown that short-term momentum persists, especially over 1 to 3 month horizons.
[17:49] Cross-sectional and relative momentum are effective at identifying outperformers and underperformers.
[17:54] And time series momentum is particularly helpful in spotting trend continuation even in volatile names like AMC.
[18:01] By incorporating these momentum signals, we can refine our spread selection.
[18:05] This boils down to favoring bullish verticals when momentum is strong and there is strong call skew and favoring bearish verticals when momentum is weak or negative and there is put skew.
[18:16] Small directional edge can meaningfully boost the already positive expectancy of a skew favored vertical spread.
[18:22] To recap, skewed vertical spreads exploit pricing inefficiencies in the implied volatility surface.
[18:29] Even without directional forecasts, these trades can offer a statistical edge due to mispricing.
[18:35] However, layering in momentum informed directional bias can enhance win rate and payoff consistency.
[18:40] To make this strategy concrete, let's walk through real vertical spread trades I've shared or that members have shared the past month with the Oquans community.
[18:48] These examples show how to use skew and momentum to construct high edge setups.
[18:52] Let's start with Qubt.
[18:56] QBT ranked positively on all major momentum metrics, cross-sectional, time series, and relative.
[19:02] This indicated strong directional follow-through to the upside.
[19:05] The call skew was not only elevated but registered a nearly -2 zed score relative to its historical average, meaning the skew was unusually steep and likely overpriced.
[19:14] The realized volatility at the time was 120%.
[19:17] Variance risk premium was approximately -7%.
[19:19] Meaning implied volatilities were slightly under realized volatility, suggesting fair pricing at the at the money and overpriced volatility at the wing.
[19:29] The trade structure that we used was to buy a June 20th expiration at the $14.50 strike at the 50 delta at an implied volatility of 113%.
[19:39] We then sold a June 20th expiration $20 call at the 10 delta at an IV of 141%.
[19:46] The total debit for this spread was $90 with a potential max profit of $460.
[19:52] The outcome was we closed near a max payout.
[19:54] We profited $345 per spread or 380%.
[19:59] The next was SBET.
[20:02] It had strong positive signals across all momentum filters.
[20:06] Despite a shorter price history, the patterns were consistent with bullish continuation.
[20:10] Extremely high call skew, several standard deviations below the mean.
[20:13] This reflected a strong lottery ticket bid on out-of-the-money calls, an ideal
[20:17] environment to sell expensive upside.
[20:19] Given the implied volatility
[20:21] differential and negative VRP, this
[20:23] trade aligned with our framework. Buy
[20:25] relatively fair implied volatility, sell
[20:27] very expensive implied volatility. The
[20:29] trade structure we used was to buy a
[20:31] July 18th expiration, $24 strike call at
[20:34] the 50 delta at an implied volatility of
[20:37] 219%.
[20:38] And sell a July 18th 32 strike call at
[20:41] the 20 delta at an implied volatility of
[20:44] 280%. The total debit for the spread was
[20:47] $140 with a potential max profit of $660
[20:50] per spread. The outcome was we closed
[20:52] for a solid win, a profit of $400 per
[20:55] spread or a return of 280%.
[20:58] The next is MP. MP showed strong
[21:00] performance across momentum categories
[21:02] with price action confirming bullish
[21:04] strength. Call skew was stretched far
[21:06] beyond historical averages offering
[21:08] excellent conditions for skew harvesting
[21:10] with realized volatility near 85 to 90%.
[21:13] The at the money option was fairly
[21:14] priced while the wing was meaningfully
[21:16] overpriced. The trade structure we used
[21:18] was to buy the July 18th expiration 49
[21:21] strike call at the 50 delta at an
[21:23] implied volatility of 86%. And sell the
[21:26] July 18th expiration 55 strike call at
[21:29] the 15 delta at an implied volatility of
[21:31] 102%.
[21:33] The total debit for the spread was $120
[21:35] with a max profit of $480. The outcome
[21:38] was that we got to close at max profit,
[21:41] $480 per spread or a 400% return. QS
[21:44] topped across all momentum factors,
[21:46] strong recent gains, outperformance
[21:48] relative to peers, and proximity to
[21:50] recent highs. Col was historically
[21:53] elevated, several standard deviations
[21:55] below its average, which was a clear
[21:57] sign of demand distortion. Realized
[21:59] volatility was comparable to implied
[22:01] volatility if we excluded earnings
[22:02] volatility since we were planning to
[22:04] exit before earnings. We put on this
[22:06] trade by buying the July 25th expiration
[22:09] 11.5 strike call at the 50 delta at an
[22:12] implied volatility of 160%.
[22:14] And we sold the July 25th expiration 16
[22:17] strike call at the 20 delta at an
[22:19] implied volatility of 200%. The total
[22:22] debit for the spread was $80 with a
[22:24] potential max profit of $370.
[22:27] And the outcome was $131 per spread or
[22:30] 160%. Let's revisit the core criteria
[22:33] that drive these trades. elevated skew
[22:35] relative to history. We can use the
[22:37] historical skew time series to identify
[22:39] when skew is unusually steep. This
[22:42] implies potential mispricings in the
[22:43] wings. We also look for momentum in the
[22:45] direction of skew. A bullish setup,
[22:48] meaning positive momentum, pairs well
[22:50] with call skew. For bearish setups, the
[22:52] logic is reversed. Use put skew and
[22:54] negative momentum. This creates trades
[22:56] where you're buying cheap volatility
[22:58] near the money and selling expensive
[23:00] skewed volatility further out. Over
[23:02] time, this edge adds up, even if the win
[23:04] rate doesn't tend to be that high. I
[23:07] generally structure these trades with
[23:09] short durations, typically 5 to 15
[23:11] calendar days. If skew remains
[23:13] overpriced over longer horizons, and
[23:15] there's conviction in sustained
[23:16] momentum, longerterm verticals can work.
[23:19] Just be prepared for a larger initial
[23:21] debit. The typical structure involves
[23:23] buying an at the money or slightly out
[23:25] of the money option likely between the
[23:27] 45 and 55 delta and selling an
[23:29] out-of-the- wing option with 10 to 25
[23:32] delta. I always price out a few
[23:34] combinations before selecting the final
[23:35] trade. The goal is to maximize reward to
[23:38] risk, skew capture, and alignment with
[23:40] momentum. Most trades are held until
[23:42] just before expiration to allow time for
[23:44] the underlying to move and the short leg
[23:46] to decay. If both options are out of the
[23:48] money near expiration, I'll usually
[23:50] allow them to expire worthless to save
[23:52] on commissions. If an earnings event is
[23:54] scheduled during the life of the spread
[23:55] and I have no directional edge or
[23:57] opinion with regards to earnings, I will
[23:59] exit the trade before earnings to avoid
[24:01] that outcome. As seen in the QS example,
[24:04] this strategy is built around a simple
[24:06] but powerful idea. Buy fair or underpric
[24:09] volatility and sell overpric skew.
[24:12] Ideally, we want this to be in the
[24:14] direction of strong momentum. We've seen
[24:16] how this played out with recent trades.
[24:18] Large reward to risk, steep skew,
[24:20] positive momentum, and favorable
[24:22] volatility conditions. Many of these
[24:24] produced tripledigit percentage returns
[24:26] on risk. But let's ground our
[24:28] expectations in reality. Remember the
[24:29] AMC example from earlier. Even though
[24:31] that trade had a strong expected return,
[24:34] the win rate was only around 32%. These
[24:36] are asymmetric trades. You lose small
[24:39] most of the time, but the wins are much
[24:41] larger and more than compensate for
[24:42] that. This means don't expect to win on
[24:45] every trade. Far from it. Stick to the
[24:48] process. The statistical edge is in the
[24:50] pricing dislocation, not in predicting
[24:52] every single move. Over time, if you're
[24:55] consistent, the math works in your
[24:56] favor. If you'd like to follow more of
[24:58] these trade setups in real time or start
[25:00] screening them yourself, check out Okans
[25:01] in the description.