SEC filings / CIK 1539638
TRIUMPH FINANCIAL, INC.
State commercial banks, TX. 12 10-K filer-years in this dataset, the most recent for FY2025, filed Feb 11, 2026. Every figure below is as filed by the company and untouched: FinObservatory has not restated, rescaled or corrected anything, and a blank is a fact the filing does not tag, not a zero.
This is the SEC holding company, identified by its Central Index Key. It is not the insured bank subsidiary that appears in /banks and /deposits, which are FDIC call-report entities on a different consolidation. Nothing here should be compared line for line with those.
Held-to-maturity gap against reported equity
The fair value of the held-to-maturity book minus its carrying amount, divided by the stockholders’ equity the same filing reports. Years marked n/a are years in which the filing does not tag one of the three.
The gap is computable for 11 of this company’s 12 filings.
Source: SEC Financial Statement Data Sets, US Securities and Exchange Commission, Division of Economic and Risk Analysis. Facts as filed: uncleaned, unrestated, not audited by FinObservatory. Selection: consolidated USD balance-sheet instants at each filing's own period end, from this company's 10-K submissions with prevrpt = 0. Methodology
Filing history, as filed
| FY | Filed | Assets | Deposits | Equity | AOCI | HTM carrying | HTM fair value | Gap, % of equity | Custom facts |
|---|---|---|---|---|---|---|---|---|---|
| 2025p | Feb 11, 2026 | $6.38bn | $4.95bn | $941.8m | -$1.7m | $1.6m | $1.9m | +0.0% | 9.8% |
| 2024 | Feb 11, 2025 | $5.95bn | $4.82bn | $890.9m | -$3.1m | $1.9m | $2.5m | +0.1% | 11.2% |
| 2023 | Feb 13, 2024 | $5.35bn | $3.98bn | $864.4m | -$2.9m | $3.0m | $4.0m | +0.1% | 8.5% |
| 2022 | Feb 15, 2023 | $5.33bn | $4.17bn | $889.0m | -$6.9m | $4.1m | $5.5m | +0.2% | 9.1% |
| 2021 | Feb 14, 2022 | $5.96bn | $4.65bn | $858.9m | $8.0m | $4.9m | $5.4m | +0.1% | 7.8% |
| 2020 | Feb 12, 2021 | $5.94bn | $4.72bn | $726.8m | $5.8m | $5.9m | $5.8m | -0.0% | 6.2% |
| 2019 | Feb 11, 2020 | $5.06bn | $3.79bn | $636.6m | $1.1m | $8.4m | $6.9m | -0.2% | 5.7% |
| 2018 | Feb 12, 2019 | $4.56bn | $3.45bn | $636.6m | -$1.2m | $8.5m | $7.3m | -0.2% | 4.1% |
| 2017 | Feb 13, 2018 | $3.50bn | $2.62bn | $391.7m | -$596k | $8.6m | $7.5m | -0.3% | 4.5% |
| 2016 | Feb 17, 2017 | $2.64bn | $2.02bn | $289.3m | -$276k | $29.4m | $30.8m | +0.5% | 3.6% |
| 2015 | Feb 26, 2016 | $1.69bn | $1.25bn | $268.0m | $277k | n/a | $0 | n/a | 3.4% |
| 2014 | Mar 6, 2015 | $1.45bn | $1.17bn | $237.5m | $951k | $745k | $750k | +0.0% | 1.1% |
ext marks a carrying amount tagged with a company-invented element. p marks FY2025, which the dataset’s filing window (closing Mar 31, 2026) leaves incomplete across filers. “Custom facts” is the share of that filing’s consolidated facts carrying a company-invented tag, in any unit.
Source: SEC Financial Statement Data Sets, US Securities and Exchange Commission, Division of Economic and Risk Analysis. Facts as filed: uncleaned, unrestated, not audited by FinObservatory. Selection: this company's 10-K submissions with prevrpt = 0, one per fiscal year by latest filed date; consolidated USD instants at each filing's period end. Dollar columns are raw US dollars. Methodology
What this page cannot tell you
- A blank in the held-to-maturity columns means the filing does not tag that fact on the consolidated face. It does not mean the company holds no held-to-maturity securities, and this dataset cannot distinguish the two.
- A held-to-maturity gap is not a realised loss. Held to maturity, the securities pay par. The gap measures how far the reported equity sits from a mark-to-market equity, which becomes a solvency question only if the book must be sold.
- AOCI is not the held-to-maturity mark. It bundles available-for-sale marks, hedge marks, pension and translation, net of tax. Do not add the two columns.
- Every value is as filed. A scale error, sign error or mis-tag by the company is reproduced here exactly as the company made it.
Pages exist for the 650 bank and thrift holding companies with at least one 10-K from FY2015 onward, out of 900 with a 10-K anywhere in the dataset.