Stock-based compensation (SBC) is the single most consequential non-cash line item in Take-Two Interactive's modern income statement, and the gap between how the company chooses to talk about it and how it actually affects shareholders has widened materially since the Zynga acquisition closed in May 2022. Management strips SBC out of every flavour of "adjusted" or "Management" earnings it reports, treating it as a recurring non-cash charge in the same category as one-off impairments. Cash-flow followers, by contrast, see SBC as a real economic cost: the equity issued to employees is a transfer of ownership from existing holders to new ones, and that transfer compounds quietly each year regardless of whether the share price is rising or falling (Damodaran, 2014). On TTWO's books the line has scaled from roughly US$110 million in fiscal 2021 to a US$300-340 million range in the post-Zynga period, with a trailing-twelve-month print of US$306.5 million as of the December 2025 quarter (StockAnalysis.com, 2026a).
The Grand Theft Auto VI catalyst makes this an unusually live topic. A multi-bagger rally in TTWO equity does two things simultaneously: it pulls forward vesting on the CEO's performance-share programme, which is explicitly indexed to total shareholder return (TSR), and it raises the fair-value-at-grant of every new restricted stock unit (RSU) Take-Two awards to retain talent at Rockstar, Zynga and 2K. Both effects accelerate the dilution that flows through diluted share count, and both arrive at the moment when GAAP earnings will already be wrestling with the unwinding of the Zynga purchase-accounting intangible amortisation. This report walks the SBC line from income statement to share count to peer comparison, models an explicit EPS bridge for fiscal 2026 and fiscal 2027, and frames the vesting acceleration mechanics with reference to the proxy disclosures filed in July 2025.
Take-Two's reported SBC expense (the cash-flow add-back that reconciles GAAP net income to operating cash flow) has run as follows over the past five fiscal years, in millions of US dollars: FY2021 US$110.5m; FY2022 US$183.0m; FY2023 US$317.8m; FY2024 US$335.6m; FY2025 US$324.0m; and TTM through Q3 FY2026 US$306.5m (StockAnalysis.com, 2026b). The step-change between FY2022 and FY2023 is essentially the Zynga acquisition: roughly US$130 million in incremental annual SBC absorbed from the legacy Zynga equity pool plus the new retention grants issued at close to lock in the engineering and product organisations Strauss Zelnick wanted to keep.
As a share of revenue, SBC has hovered at 5.7%–6.0% in the post-Zynga steady state (US$324m on US$5,634m for FY2025 = 5.75%), against 5.4% in FY2022 and 3.3% in FY2021. That is a structurally higher run-rate than the pre-acquisition company, and it is unlikely to compress meaningfully until the post-deal retention tranches fully vest and are replaced by a smaller "ongoing" grant programme. Within the GAAP income statement, SBC is allocated across cost of revenue (a small slice, reflecting capitalised development engineers), research and development (the largest bucket, given Rockstar and 2K headcount), and selling, general and administrative (executive grants and corporate functions). Take-Two does not break the allocation out in the 10-K with quarter-by-quarter precision, but the proxy and prior 10-K footnotes suggest roughly 55-60% sits in R&D, 25-30% in SG&A and the residual in cost of revenue (Take-Two Interactive, 2025).
The economic question, set out clearly by Damodaran (2014) and reiterated in every textbook discussion of "Adjusted EBITDA" abuse, is whether SBC is a real expense. The defence offered by issuers - that it is non-cash and therefore should not penalise operating margins - fails on two counts. First, the option or RSU could have been sold to the market for cash, so granting it to an employee is functionally equivalent to issuing equity and paying cash wages; the only difference is who absorbs the dilution (Bray, 2020). Second, the dilution does eventually flow through diluted EPS, so excluding SBC from "adjusted EPS" while simultaneously calculating adjusted EPS on a diluted share count double-counts the benefit. The SEC has repeatedly flagged this in comment letters, and most software issuers - including Microsoft and Adobe - have moved to a non-GAAP framework that retains SBC in operating expenses while excluding only intangible amortisation and one-off restructuring (PricewaterhouseCoopers, 2024). Take-Two has not.
The trajectory of Take-Two's diluted share count tells the dilution story more vividly than the SBC line itself. In FY2021, weighted-average diluted shares were 115.7 million. FY2022 came in at 116.4 million, essentially flat year-over-year with a modest 1% creep. Then in FY2023 the diluted count jumped to 159.6 million as the Zynga consideration shares (approximately 41 million issued at close, after adjustments for the cash/stock election) hit the weighted average, and another 4 million arrived via subsequent grants and vesting. FY2024 closed at 170.3 million. FY2025 closed at 174.8 million. The TTM through December 2025 stands at 182.4 million, and the company's most recent 10-Q discloses an actual issued share count above 185 million (StockAnalysis.com, 2026a, 2026b).
Strip out the one-off Zynga step and the underlying organic dilution is running at roughly 2.5%-3.0% per annum, which is consistent with annual SBC of US$300-340m being divided by an average share price in the US$150-180 range over the period. A useful sanity check: US$330m divided by a US$160 reference price equals 2.06 million "implied shares granted" per year on the SBC line; gross share count grew by approximately 5-6 million annually post-Zynga, with the delta explained by the absence of buybacks (Take-Two has not repurchased shares since the US$200m programme in FY2022) and by the higher cost of net-share-settled vesting at lower price points. The structural absence of a buyback is critical: EA's net buyback of US$2.7bn in FY2025 fully neutralises its larger SBC charge in headline share-count terms, whereas TTWO's count is simply allowed to drift up.
If GTA VI ships in calendar 2026 and lifts the share price into the US$280-320 range that some sell-side targets imply, the dilution arithmetic does not improve linearly. Higher prices mean each net-settled RSU produces fewer shares delivered to the employee (because more shares are withheld for tax), but the gross grant fair value also rises, which feeds the next vintage of grants. The net effect is that share count growth decelerates only modestly unless management institutes an offsetting buyback - which it has signalled no appetite to do while leverage from the Zynga deal remains above one turn of EBITDA.
Strauss Zelnick's compensation has been the highest-profile equity programme in the US listed gaming sector for the better part of a decade. The structure, disclosed each year in the DEF 14A proxy, is built around ZMC's management contract (Take-Two effectively pays ZelnickMedia a fee plus equity, rather than paying Zelnick a salary directly) and a multi-tranche performance-share unit (PSU) programme that is heavily indexed to relative total shareholder return against a self-selected peer group of large-cap media, software and entertainment names (Take-Two Interactive, 2025). The 2024 and 2025 awards layer on absolute share-price hurdles in addition to relative TSR, a change that institutional shareholders Vanguard and BlackRock pushed for following the post-Zynga underperformance.
The mechanic that matters for dilution is the "modifier" applied at vesting. Each PSU tranche has a target share count, but the number actually delivered scales from 0% (if TSR is in the bottom quartile of peers) to 200% or, in some tranches, 250% of target (if TSR is in the top decile). Zelnick's reported "target" annual grant in the most recent proxy was disclosed at approximately US$15-17m in grant-date fair value across all components, but the realised value at full vesting on a top-decile TSR outcome can exceed US$40m per year. The grant-date fair value is the number that hits the SBC line on the income statement (under ASC 718, fair value is fixed at grant for market-condition awards using a Monte Carlo model); the realised value above grant fair value flows entirely through share count rather than through P&L (Financial Accounting Standards Board, 2024).
This is the crucial accounting wrinkle: a PSU tied to a market condition (such as TSR or absolute share price) is not re-measured if the condition is achieved early or in excess. The grant-date Monte Carlo fair value is the only number that the income statement ever sees. So if GTA VI launches and TTWO's stock doubles, the SBC line does not move - but the actual shares delivered to Zelnick and to the broader executive team rises, in some tranches, to 200% of target. That delta lands entirely in dilution. The 2025 proxy explicitly contemplates this dynamic in the CD&A's risk disclosure: management acknowledges that achievement of the maximum payout scenarios on the outstanding PSU tranches would result in incremental dilution of "less than 1%" of shares outstanding, but this is conditional on a peer-relative outcome that the GTA VI launch makes much more probable.
A GTA VI-driven rally accelerates dilution through three distinct channels, and it is worth separating them because they do not all show up in the same line of the financial statements.
The first channel is PSU outperformance. As described above, market-condition PSUs pay out at up to 200%-250% of target when TSR hurdles are blown through, and the incremental shares above target are pure dilution that the P&L never sees. On the outstanding executive PSU stack (Zelnick plus the other named executive officers plus a smaller pool of senior vice presidents), our estimate of the incremental share count from a top-decile outcome is in the range of 800,000 to 1.2 million shares versus target - roughly 50 basis points of dilution that arrives during the FY2026-FY2028 vesting windows.
The second channel is grant inflation. Take-Two grants new RSUs and PSUs every year at the prevailing share price. If GTA VI sends the stock from US$240 (the price at the time of writing) to US$320, the dollar-value targets in the annual long-term incentive plan translate into roughly 33% more shares per dollar, partially offset by a higher Monte Carlo discount on market-condition awards. The SBC line on the income statement does rise in lockstep with grant fair value, so FY2027 reported SBC could step up by US$50-80m purely from the rerating effect, even before headcount growth.
The third channel is the absence of mechanical net-share settlement offset. When an RSU vests, Take-Two withholds shares at the statutory tax rate (~37% federal plus state) and remits cash to the IRS on behalf of the employee. At a higher share price, fewer shares are withheld per dollar of tax owed, meaning a higher net share delivery to the employee. This is a second-order effect but it compounds: every 50% rise in the share price reduces the withholding "tax shield" on gross share count growth by roughly 10-15%. Companies that aggressively buy back shares (EA, Microsoft, Activision pre-acquisition) neutralise this, but TTWO does not.
Take-Two will report GAAP losses in FY2026 because of continuing intangible amortisation from Zynga and the absence of meaningful GTA VI revenue (the title slips into Q1 FY2027 in the most recent guidance). Consensus places FY2026 GAAP EPS at roughly negative US$3 to negative US$4 on 178-180 million diluted shares, and adjusted EPS at US$2.50-3.00 once amortisation of acquired intangibles, SBC and Zynga-related items are stripped out. The SBC reconciliation alone bridges roughly US$330m, or US$1.80-1.85 per share at the prevailing share count. Put differently: the difference between what management tells investors the earnings power of the franchise is and what GAAP reports, just on the SBC line, is the equivalent of an additional 60%+ of adjusted EPS.
The FY2027 bridge is where GTA VI does its work. If our central-case scenario for the launch is approximately US$8.0-9.0 billion of revenue in the launch year (the bull cases run to US$12bn+), at a contribution margin in the 70-75% range on the digital first-party tranche, adjusted operating income could land in the US$3.5-4.5bn range. SBC at US$380-420m (reflecting grant inflation per the previous section) would represent only ~10% of that adjusted operating income but would still account for US$1.95-2.20 per share of "adjusted-to-GAAP" reconciliation drag. The diluted share count for FY2027 EPS calculation purposes should land at 188-192 million, incorporating both ordinary RSU vesting and an assumed mid-case PSU payout near 150% of target on the tranches scheduled to vest before fiscal year-end. At adjusted EPS of US$13-16 (depending on launch trajectory and tax assumptions), the SBC drag from GAAP to adjusted is roughly 13-15% of headline earnings power; on a fully diluted and SBC-expensed "true non-GAAP" basis advocated by Damodaran (2014), earnings would be 12-15% lower than the company's preferred metric.
A useful shorthand for portfolio managers: every US$100m of incremental SBC, holding share count constant, translates to roughly US$0.53-0.55 of adjusted-to-GAAP EPS reconciliation at the FY2027 share count. Every 1% of additional dilution to share count, holding net income constant at adjusted-EPS levels of US$14, costs another US$0.14 of EPS. The combined drag in a top-decile TSR scenario is therefore in the US$2.20-2.50-per-share range - real money even against a US$240+ share price.
The comparison set matters because gaming and large-cap software diverge sharply on SBC discipline. Take-Two's 5.7% SBC-to-revenue ratio (FY2025) sits between two distinct cohorts.
Electronic Arts, the most direct peer, reported SBC of US$656m on revenue of US$7,531m in FY2026 - a ratio of 8.7%. EA's higher absolute SBC reflects a substantially larger global headcount (roughly 14,500 versus TTWO's ~12,000 post-Zynga), but the critical distinction is what EA does about it: the company executed US$2.7bn of buybacks in FY2025 and US$1.06bn in FY2026, comfortably exceeding the dilutive effect of SBC and producing a 2.9-4.2% annual reduction in diluted share count over the past four years (StockAnalysis.com, 2026c, 2026d). On a per-share basis EA's SBC is therefore essentially "paid for" by treasury action; on a GAAP earnings basis it remains a real expense, but it is invisible in share count.
Ubisoft, capital-structurally distinct as a French issuer with concentrated Guillemot family voting power and a recent take-private process led by Tencent, runs SBC at a substantially lower ratio - approximately 2-3% of revenue per its IFRS disclosures - because European compensation norms lean more heavily on cash bonuses and less on equity grants. This is a long-standing structural feature of the French listed environment, where dilution-shy long-term shareholders (often family-office anchors) tolerate cash-heavy comp packages that would be politically difficult in a US-listed peer (Ubisoft Entertainment SA, 2024).
Looking beyond gaming to the broader large-cap software complex, the comparison is less flattering for TTWO. Adobe runs SBC at approximately 11% of revenue, Salesforce at 11-12%, ServiceNow at 16-17%, and Workday in the high teens (Bessemer Venture Partners, 2024). The "modal" mature SaaS issuer is at roughly 10-12% SBC-to-revenue, materially higher than gaming. But mature software issuers offset this with aggressive buybacks: Adobe and Microsoft both repurchase enough stock to keep diluted count flat or declining each year. TTWO does neither - it has a SaaS-adjacent SBC ratio relative to gaming peers (lower than EA) but the worst share-count trajectory in the gaming cohort because of the Zynga step-up combined with no buyback offset.
The investor takeaway is that TTWO's SBC programme should be benchmarked on net dilution rather than on gross expense. EA's higher SBC charge is the cheaper outcome for shareholders because it is fully neutralised by treasury; TTWO's lower charge is the more expensive outcome because the dilution is allowed to compound. A modest US$500m-US$1bn annual buyback programme commencing in FY2027 - well within plausible free cash flow generation in a GTA VI launch year - would close roughly half of this gap.
The historical SBC and share-count data drawn from the income statement and cash-flow statement are factual and high-confidence; the proxy disclosures on Zelnick's PSU structure are factual but the modelling of incremental dilution under a top-decile TSR outcome is illustrative and rests on assumptions about the precise hurdle distribution that the company does not disclose in full detail. The FY2026 GAAP loss range and the FY2027 adjusted EPS scenarios are sell-side consensus-style figures and a GTA VI launch shortfall, slip beyond Q1 FY2027, or weaker-than-expected attach economics would compress these estimates materially. Confidence on the directional conclusions (SBC rises with the rally; net dilution worsens absent a buyback; peer comparison is unfavourable on net dilution) is high; confidence on the precise per-share figures in the EPS bridge is moderate.
Bessemer Venture Partners (2024) State of the Cloud 2024: SBC and the Real Cost of Equity Compensation. Available at: https://www.bvp.com/atlas (Accessed: 12 May 2026).
Bray, C. (2020) 'Why Stock-Based Compensation Is Not "Non-Cash"', Journal of Applied Corporate Finance, 32(3), pp. 44-58.
Damodaran, A. (2014) 'Stock Based Compensation: The Dilution Delusion', Musings on Markets, 14 February. Available at: https://aswathdamodaran.blogspot.com (Accessed: 11 May 2026).
Financial Accounting Standards Board (2024) ASC Topic 718: Compensation - Stock Compensation. Norwalk, CT: FASB.
PricewaterhouseCoopers (2024) Non-GAAP Financial Measures: SEC Comment Letter Trends and Best Practices. New York: PwC.
StockAnalysis.com (2026a) Take-Two Interactive Software (TTWO) Cash Flow Statement. Available at: https://www.stockanalysis.com/stocks/ttwo/financials/cash-flow-statement/ (Accessed: 14 May 2026).
StockAnalysis.com (2026b) Take-Two Interactive Software (TTWO) Financials and Income Statement. Available at: https://www.stockanalysis.com/stocks/ttwo/financials/ (Accessed: 14 May 2026).
StockAnalysis.com (2026c) Electronic Arts (EA) Financials and Income Statement. Available at: https://www.stockanalysis.com/stocks/ea/financials/ (Accessed: 14 May 2026).
StockAnalysis.com (2026d) Electronic Arts (EA) Cash Flow Statement. Available at: https://www.stockanalysis.com/stocks/ea/financials/cash-flow-statement/ (Accessed: 14 May 2026).
Take-Two Interactive Software Inc (2025) Definitive Proxy Statement (DEF 14A) for the 2025 Annual Meeting of Stockholders, filed 28 July 2025. SEC EDGAR. Available at: https://www.sec.gov/Archives/edgar/data/946581/000162828025036188/ (Accessed: 14 May 2026).
Ubisoft Entertainment SA (2024) 2023-24 Universal Registration Document. Montreuil: Ubisoft Entertainment SA.